Menière's Syndrome Associated With Fibrous Dysplasia of the Temporal Bone.

BACKGROUND Pathologies of the endolymphatic sac (ES) most likely diminish the inner ear's capability to maintain normal ion and fluid homeostasis and, more recently, have been systematically linked to the pathogenesis of Menière's symptoms (1,2). Different ES pathologies have been identified among patients with Menière's symptoms; the ES either fails to develop correctly (ES hypoplasia) (1), progressively degenerates after initial normal development (ES degeneration) (1), or becomes secondarily affected by various disease mechanisms originating in the posterior temporal bone or the posterior cranial fossa (2,3). Here, we report for the first time on fibrous dysplasia (FD) of the posterior temporal bone that erodes the ES and is clinically associated with Menière's symptoms. CASE PRESENTATION A 41-year-old woman was referred to our department with a 1-year history of episodic vertigo, with each symptom episode lasting several hours, occurring two to three times a week with no identifiable triggers, and being accompanied by left-sided hearing loss, tinnitus, and aural fullness. Moreover, she had suffered from several episodes of transient low-frequency sensorineural hearing loss in her left ear. Her medical history otherwise included polyostotic FD, which had been first diagnosed in her early childhood, and presented with foci in the left femur, left tibia, and several cranial bones. She reported that she had received an endocrinologic workup, which had not resulted in a syndrome diagnosis, nor had it shown any other organ involvement. Initial audiovestibular testing demonstrated moderate to severe sensorineural hearing loss on the left and normal hearing on the right (Fig. 1A), left-sided caloric hyporeflexia, and a reduced vestibulo-ocular reflex gain (0.59; Fig. 1B). High-resolution computed tomography (HRCT) and magnetic resonance imaging (MRI) revealed an FD focus affecting the petrous apex and the retrolabyrinthine area (posterior temporal bone), which eroded and obliterated the vestibular aqueduct (VA; Figs. 1, C–E, and 2, A and B). Notably, the FD focus did invade neither the otic capsule bone, nor the inner ear, nor the internal auditory canal (Figs. 1, C–E, and 2, A–F). Noncontrast T1- and T2-weighted MRI showed within the FD focus an unspecific, heterogenous low to intermediate signal intensity (Fig. 1D). Gadolinium-enhanced T1-weighted MRI revealed a focal area of increased gadolinium enhancement in the posterior temporal bone, indicating a particularly active bone remodeling zone (4) (Fig. 1E). The 4-hour delayed (post gadolinium injection) three-dimensional inversion recovery sequence, used to visualize the inner ear fluid spaces, revealed a left-sided endolymphatic hydrops (Fig. 2, C and D). The inversion recovery signal of the VA was interrupted within the FD focus, corresponding to the bony obliteration seen on HRCT (Fig. 2, A–F). Imaging overall did not reveal other retrolabyrinthine or intralabyrinthine pathologies. Based on the aforementioned findings, a diagnosis of left-sided Menière's syndrome was made according to established diagnostic criteria (5). After initial treatment with betahistine up to 72 mg three times daily for 6 weeks and three serial intratympanic (i.t.) dexamethasone injections (4 mg/ml; 0.3 ml/injection), which both did not improve the frequent vertigo symptoms, treatment was escalated to a single i.t. injection with low-dose (4 mg) gentamicin. Over the following weeks, the frequency and intensity of vertigo symptoms gradually decreased, and after 6 weeks, the vertigo attacks eventually ceased. Follow-up audiovestibular testing, 4 months after initial presentation, showed unchanged hearing thresholds (left pure-tone average improved by 4 dB) and a vestibulo-ocular reflex gain reduction to 0.33. The patient did not report any vertigo attacks during a follow-up of 6 months.FIG. 1: A and B, Audiovestibular testing. A, Pure-tone audiometry revealing a moderate to severe sensorineural hearing loss in the left ear (pure-tone average 0.5, 1, 2, and 4 kHz: 64 dB) and age-adjusted normal hearing thresholds on the right. AC, air conduction; BC (mask), bone conduction (masked). B, Video head impulse test of the left lateral canal showing a reduced vestibulo-ocular gain of 0.59 and scattered catch-up saccades (square bracket). Blue, head movement traces; black, eye movement traces. C, Axial noncontrast HRCT revealed an expansive bone lesion with ground-glass sclerotic appearance affecting the left temporal bone (asterisks) and the clivus. The lesion is relatively homogeneous and has a distinct cortical margin, characteristic of FD. Note the unaffected, dense bone of the otic capsule surrounding the cochlea (Co) and the vestibular labyrinth. Bar plot shows mean radiodensity (expressed in HU) of the otic capsule and retrolabyrinthine area (measured at the black dotted circles). A significantly lower radiodensity of the posterior temporal bone was found on the affected side compared with the unaffected side (unpaired t test). By contrast, the radiodensity of the otic capsule was not significantly different between the two temporal bones. Whiskers indicate standard deviation. Scale bar (for C–E): 1 cm. D, Axial noncontrast T1-weighted MRI showed an unspecific, heterogenous low to intermediate signal intensity of the temporal bone lesion (asterisks). E, Axial gadolinium-enhanced T1-weighted MRI revealed a focal area of increased gadolinium enhancement (arrowheads). FD, fibrous dysplasia; HRCT, high-resolution computed tomography; HU, Houndsfield units; Ma, mastoid; MRI, magnetic resonance imaging.FIG. 2: Imaging of the VA and the endolymphatic space. A and B, On axial HRCT, the VA (empty arrowhead) was well visible on the unaffected right side (A). In contrast, the VA on the affected side (B) was eroded by FD (dotted line, assumed previous course of the VA; arrow, posterior semicircular canal). Scale bar (for A–F), 5 mm. C and D, On axial delayed intravenous contrast-enhanced 3D IR MRI, Gd enhancement corresponding to the VA (empty arrowhead) was found on the unaffected side (C). No endolymphatic hydrops was present neither in the cochlea nor in the vestibular labyrinth. On the affected side (D), endolymphatic hydrops (Baráth grade I) was found in the cochlea (arrow heads) and in the vestibule (×). No Gd signal corresponding to the VA was detected (dotted line, assumed previous course of the VA; arrow, posterior semicircular canal). E and F, Overlay of HRCT and 3D IR Gd-MRI (pseudo-color image) shows Gd enhancement corresponding to the VA (empty arrowhead) on the unaffected side (E). On the affected side (F), there was no enhancement in the area of the VA (red dotted line). The pseudo-color gradient was generated using Adobe Photoshop (version 24.0.0; Adobe, San Jose, CA) using a predefined color spectrum ranging from blue to red coding for the spectrum ranging from black to white representing the 3D IR Gd-MRI signal intensity. As a reference, the original images are shown in panels C and D. 3D IR, three-dimensional inversion recovery; FD, fibrous dysplasia; Gd, gadolinium; HRCT, high-resolution computed tomography; MRI, magnetic resonance imaging; VA, vestibular aqueduct.DISCUSSION FD is a rare genetic bone metabolic disorder characterized by abnormal fibro-osseous proliferation, affecting one (monostotic form) or multiple bones (polyostotic form). The temporal bone is rarely affected in FD; the cases reported in the literature commonly presented either with stenotic external or internal auditory canals, manifesting clinically as otitis media, cholesteatoma, or cranial nerve (VII, VIII) compression syndrome (6). For the present case, in which the otic capsule bone appeared unaffected by FD based on HRCT imaging, we propose the following pathophysiology to explain the presence of endolymphatic hydrops and Menière's symptoms: bony remodeling of the retrolabyrinthine portion of the temporal bone encroached and obliterated the VA, thereby separating the ES from the labyrinth. This loss of ES function supposedly disturbed the inner ear's fluid and ion balancing capacity and most likely caused endolymphatic hydrops and fluctuating Menière's symptoms. It is noteworthy that this proposed pathomechanism parallels that in other cases in which Menière's syndrome was associated with a bony obliterated VA—either due to otosclerotic remodeling of the VA or by callus bone formation within a VA-violating temporal bone fracture (2). Bony VA malformations and accompanying ES pathologies were also demonstrated in approximately 30% of patients with idiopathic Menière's disease (1). Notably, treatment of refractory vertigo in the present case was escalated from i.t. dexamethasone directly to i.t. gentamicin, skipping ES surgery as an optional intermediate step, as it is proposed by various clinical management algorithms. The rationale against ES decompression surgery was based on the HRCT finding of the bony obliterated VA, which made it unlikely that surgically targeting the distal (disconnected) ES may have any effect on the patient's vestibular symptoms. An open question concerns the time interval between bony VA obliteration and the onset of Menière's syndrome in the present case. The observed decades-long time interval to the onset of Menière's symptoms in patients with a—likely—hereditary VA malformation (1) or a VA-violating fracture (2) may suggest a similarly long asymptomatic manifestation of the temporal bone FD focus in the present case. In summary, the present case (i) adds to the growing evidence that a selective compromise (loss) of the ES is critical in the pathogenesis of endolymphatic hydrops and Menière's symptoms, (ii) establishes FD of the posterior temporal bone as a new potential cause for secondary Menière's syndrome, (iii) emphasizes the importance for scrutinizing the posterior temporal bone in HRCT imaging for potential VA pathologies in cases presenting with Menière's symptoms, and (iv) illustrates that therapeutic procedures targeting the ES should not be considered if imaging suggests a discontinuity between the ES and the cochleovestibular labyrinth, such as a bony obliterated VA.