Unusual MRI appearance of an intracranial cholesteatoma extension: the 'billiard pocket sign'. (Original Article).

Abstract

We describe a unique case of an intracranial extension of acquired cholesteatoma. Previous reports have described cholesteatoma extension through the middle fossa plate and into the middle cranial fossa, but to our knowledge ours is the first report of a case in which the sac herniated into the temporal lobe and overlying dura from a site far lateral to the otic capsule. The findings on magnetic resonance imaging were most unusual, and we call the radiologic characteristics of the mass in this case the "billiard pocket sign." We also discuss the possible mechanisms that produced such an image.

Introduction

Intracranial extension of an acquired cholesteatoma is a rare complication of unsafe atticoantral chronic suppurative otitis media. Congenital and acquired cholesteatomas can spread intracranially from the temporal bone into either the middle or the posterior fossa. The fundamental steps involved in intracranial spread have been reported to be (1) the entrapment of the cholesteatoma in a narrow recess, (2) erosion of the bone of the dural plates, and (3) expansion of the cholesteatoma into the cranial fossae. (1) Until now, there has been no report of an intracranial extension of a cholesteatoma that appeared on magnetic resonance imaging (MRI) as a spherical mass with two components of different intensities. We report such an appearance, which we call the "billiard pocket sign." We discuss the pathogenetic, diagnostic, and surgical features of this case.

Case report

A 62-year-old man was referred to our unit with a 6-year history of right-sided hearing loss and offensive otorrhea. He was known to have a cholesteatoma that arose from a posterior mesotympanic retraction pocket. One year earlier, the referring surgeon had performed a canal-wall-down tympanoplasty. During this operation, the surgeon noted a large cholesteatoma that extended over the middle fossa dura and into the middle cranial fossa through a defect in the middle fossa plate. The bone over the sigmoid sinus was eroded, the tympanic portion of the facial nerve was exposed, and the incus was eroded. Following a review of the patient in the office, the referring surgeon came to believe that the eradication of the cholesteatoma had been incomplete, and he transferred the care of the patient to our department.

During our evaluation, the patient complained of a right-sided hearing loss, high-pitched right tinnitus, unsteadiness, constant generalized headache, and right mastoid ache. Otoscopy revealed that the mastoid cavity was dry and the neotympanum was well healed. The patient's facial nerve function was normal (House-Brackmann grade I). Pure-tone audiometry at 0.5, 1, 2, and 4 kHz yielded an air-conduction pure-tone average of 66.25 dB and a bone-conduction average of 20 dB. Findings in the left ear were normal.

High-resolution axial and coronal computed tomography (CT) showed a mass of soft-tissue density in the middle ear and erosion of the tegmen tympani and middle fossa plates lateral to the lateral semicircular canal (figure 1). A subsequent axial-view T1-weighted MRI demonstrated that the spherical lesion had indented the temporal lobe of the brain (figure 2). The sac consisted of two components, and their appearance resembled that of a billiard ball in a pocket. The fundus was of high signal intensity. On coronal-view T1-weighted MRI, the more proximal portion of the lesion at the neck was of low signal intensity (figure 3). On T2-weighted MRI, both components of the sac were of high signal intensity; they did not enhance with gadolinium DTPA (diethylene-triamine-penta-acetic acid) contrast. The lining of the roof of the mastoid cavity was of high signal intensity on T1-weighted imaging.

The patient underwent surgery via a transmastoid-middle fossa approach. Intraoperatively, we noted that the middle fossa dura lateral to the labyrinth was indented and that the cholesteatoma matrix had penetrated it. We divided the temporal dura and gently retracted the temporal lobe. We noted that the sac was adherent to the tegmental dura, but no intradural extension was evident. We then performed a subtotal petrosectomy and a blind sac closure of the external auditory canal. We removed the sac and the surrounding dura and reconstructed the area with fascia lata. Finally, we closed the mesotympanic end of the eustachian tube and obliterated the middle ear and mastoid with fat.

Histologic examination revealed that the distal portion of the sac was made up of cholesterol crystals, hemosiderin-laden macrophages, and foreign-body giant cells. The proximal portion of the sac consisted of a cholesteatoma. The patient's postoperative course was uneventful.

Two years after surgery, the patient had a residual conductive hearing loss of 60 dB on the right and his bone conduction thresholds were unchanged. His facial nerve function remained normal. MRI detected no sign that the cholesteatoma had recurred.

Discussion

Intracranial extension is a recognized complication of tympanomastoid cholesteatoma. (2) Extension can occur as a result of an erosion of the tegmen tympani and occasionally iatrogenically. Cholesteatomatous extension into the middle cranial fossa usually spreads like a sheet over the dura. (1) To the best of our knowledge, sac herniation into the temporal lobe and overlying dura has not been previously described.

In our patient, the radiologic, surgical, and histologic findings suggested that there were two components to the cholesteatomatous extension. The outer part of the mass was hyperintense on T1- and T2-weighted MRI, which is a sign of cholesterol granuloma. The central part of the lesion was hypointense on T1-weighted imaging and hyperintense on T2-weighted imaging, a finding that is consistent with a cholesteatoma. (3,4)

Whether the intracranial extension of the cholesteatoma in our patient was iatrogenic or attributable to tegmental erosion is difficult to discern. Recently, Horn extensively reviewed intracranial extension of acquired cholesteatomas and noted that their spread into the middle cranial fossa occurs mainly through the supratubal recess along the labyrinthine facial nerve or above the internal auditory canal. (1) The intracranial extension of a cholesteatoma located lateral to the labyrinth, as occurred in our patient, has not been previously reported. The factors necessary for the development of a cholesterol granuloma are hemorrhage, interference with clearance or drainage, and obstruction of air exchange or ventilation. (5) The rupture of a cholesteatoma sac and the spillage of keratinous substance into the subepithelial layer usually causes a marked granulomatous reaction of a foreign-body type, but it is not associated with cholesterol crystals and hemosiderin-laden macrophages. (5) These facts would make i t unlikely that the intracranial extension in our patient was related to the tegmental erosion.

On the other hand, several aspects of our case support the premise that the extension was iatrogenic--a complication of his earlier surgery. The low-lying middle fossa dura, which was located at the level of the lateral semicircular canal (figure 1), might have rendered the mastoidectomy more complex and thereby increased the risk of surgical trauma to the dura. The middle fossa dura could have been damaged by the drill, and this might have caused both the indentation of the dura and the hemorrhage responsible for the formation of the cholesterol granuloma.

On MRI, cholesteatomas are hypointense on TI-weighted images and hyperintense on T2 images; by contrast, cholesterol granulomas are hyperintense on both TI and T2 MRIs. (3,4) However, it is possible that the high signal intensity on T1 imaging in our patient was related to the hemorrhage, and the low signal intensity might have represented artifact from bits of metal left behind during the drilling. (6)

High-resolution CT is the imaging modality of choice for evaluating tympanomastoid cholesteatomas. Bony-window CT studies are unsurpassed in demonstrating secondary bone erosion. When evidence on CT suggests invasion into the middle or posterior cranial fossa, further imaging is required before surgery can be undertaken in order to exclude intracranial extension. MRI offers the superior soft-tissue differentiation that is necessary to distinguish between cholesteatoma, cholesterol granuloma, and intracranial suppurative complications. (3,4) MRI has an important complementary role to high-resolution CT in those patients who have cholesteatoma with extensive erosion of the temporal bone and possible intracranial extension; in these cases, soft-tissue differentiation is essential.

Surgical treatment is aimed at the complete removal of disease. The combined transmastoid-middle fossa approach allowed us to completely excise the disease in our patient. The large area of dural involvement and the status of the contralateral ear led us to perform a subtotal petrosectomy and obliteration of the middle ear and mastoid cleft on the affected ear. (7) The cochlear reserve in the operated ear provides an opportunity for further rehabilitation with amplification.

References

(1.) Horn KL. Intracranial extension of acquired aural cholesteatoma. Laryngoscope 2000;110:761-72.

(2.) Burggraaff B, Luxford WM, Doyle KJ. Neurotologic treatment of acquired cholesteatoma. Am J Otol 1995;16:480-5.

(3.) Mafee MF, Kumar A, Heffner DK. Epidermoid cyst (cholesteatoma) and cholesterol granuloma of the temporal bone and epidermoid cysts affecting the brain. Neuroimaging Clin N Am 1994;4:561-78.

(4.) Robert Y, Carcasset S, Rocourt N. et al. Congenital cholesteatoma of the temporal bone: MR findings and comparison with CT. AJNR Am J Neuroradiol 1995;16:755-61.

(5.) Ferlito A, Devaney KO, Rinaldo A, et al. Clinicopathological consultation. Ear cholesteatoma versus cholesterol granuloma. Ann Otol Rhinol Laryngol 1997;106:79-85.

(6.) Sridhar K, Ramamurthi R, Vasudevan MC, Ramamurthi B. Magnetic resonance imaging artifact following acoustic neurofibroma surgery--case report. Neurol Med Chir(Tokyo) 1999;39:938-40.

(7.) Atlas MD, Moffat DA, Hardy DG. Petrous apex cholesteatoma: Diagnostic and treatment dilemmas. Laryngoscope 1992; 102:1363-8.

From the Department of Otoneurological and Skull Base Surgery, Addenbrooke's Hospital, Cambridge, U.K.

Reprint requests: David A. Moffat, FRCS, Department of Otoneurological and Skull Base Surgery, Addenbrooke's Hospital, Cambridge, UK. Phone: [phone]; fax: [phone]; e-mail: [e-mail]