Osteoid Osteoma Invading the Posterior Labyrinth of the Petrous Bone

Discussion

OO is a frequent, benign, primary bone tumor accounting for approximately 10% of all primary bone tumors.^1,2 It is most frequently seen in teenagers and young adults and is very rare after 30 years of age. Although our patient was a woman, OO usually prevails in men with a sex ratio of 2:1. Moreover, the location of the lesion in the inner ear is unique because OO mostly involves long bone diaphysis, extremities, and the vertebral column. Temporal bone OO has previously been reported in the mastoid and the middle ear^3–5 but, to the best of our knowledge, never in the labyrinth.

Pain is the main presenting symptom of OO. It usually increases during the night and is relieved by salicylates. The lack of pain with OO is uncommon. It was reported in only 1.6% in a large review⁶ and was attributed to the lack of a hard shell or nerve endings around the lesion. In our case, the lack of nerve endings in the otic capsule could explain this unusual clinical feature.⁷ Vertigo, tinnitus, and hearing loss described in our patient might be related to the particular location of the tumor. Indeed, these symptoms may be related to the posterior labyrinth involvement and especially the endolymphatic duct.

On CT, OOs typically demonstrate fusiform sclerotic cortical thickening in the shaft of a long bone. A characteristic radiolucent area, measuring <1 cm in diameter and representing the lesion itself, is usually located within the center of the area of osteosclerosis and harbors central calcification in approximately 50% of cases.^2,5,8 The lesion in our patient showed such a radiolucent area, with partial central calcification, but no peripheral sclerotic bone was clearly conspicuous because it may have been lost in the high attenuation of the otic capsule.

The presence of a radiolucent lesion with internal calcifications and close contact with the facial nerve also raises the possibility of a temporal bone hemangioma. Although the topography was very unusual for the latter (hemangioma is usually centered near the geniculate ganglion or the internal auditory canal) and no facial nerve palsy was observed, the overall imaging pattern is almost indistinguishable between the entities and final diagnosis is determined by histology.

At MR imaging, the lesion demonstrated intermediate signal intensity on T1-weighted images and high signal intensity on T2-weighted images and heterogeneously enhanced after gadolinium injection. Endolymphatic sac tumors may involve this region and appear as a heterogeneously enhancing osteolytic tumors arising from the endolymphatic duct. However, endolymphatic sac tumors usually erode the posterior aspect of the petrous pyramid and extend into the cerebellopontine angle cistern. Additionally, at MR imaging, endolymphatic sac tumors are heterogeneous, demonstrating areas of T1 and T2 high signal intensities corresponding to cysts filled with blood or high proteinaceous fluid and multiple small signal intensity voids due to calcifications and blood vessels.⁹

Other considerations for enhancing petrous bone lesions include meningioma, eosinophilic granuloma, glomus tumor, cartilaginous tumors, chordoma, and metastasis, but these are usually larger when symptomatic.¹⁰ The size of the lesion is an important criterion because it also helps in distinguishing OO from osteoblastoma, another osteoid-producing tumor. This distinction can be very challenging, even on histology¹¹ because both lesions harbor a hypervascular young conjunctival stroma, which contains osteoblasts and multinucleated giant cells in thin osteoid rows that can be calcified.^2,4 In equivocal cases, the size of the lesion measured on CT remains the most important criterion: OOs are typically <1 cm, whereas osteoblastomas are >2 cm.¹¹ Another finding that can distinguish the 2 entities in theory is central calcification, because OO usually contains a single calcification, whereas an osteoblastoma contains multiple calcifications.¹¹ Therefore, our case could have been labeled osteoblastoma, but it did not meet the size criteria. In fact, all osteoblastomas of the temporal bone reported in the English literature were noticeably larger than 2 cm.¹²

It is important to differentiate OO from osteoblastoma because they have different behaviors and treatments. Generally regarded as “benign,” osteoblastomas have definite potential for aggressive behavior, local recurrence, and malignant transformation, while OOs do not.