‘Cochlear-type’ hearing loss as part of aquaporin-4 neuromyelitis optica spectrum disorder

Abstract

Neuromyelitis optica spectrum disorder is an inflammatory autoimmune central nervous system condition caused in the majority of cases by aquaporin-4 IgG antibodies. Aquaporin-4 is expressed in the cochlear and vestibular nuclei regions in the brainstem and a handful of cases of retro-cochlear type hearing loss have been documented in the literature. Aquaporin-4 has also been reported within the organ of Corti and the cristae and maculae of the vestibular apparatus. We present a case where there is evidence of peripheral (labyrinthine) rather than central pathology and propose this is due to autoimmune inflammation as part of neuromyelitis optica spectrum disorder. This is the first case in the literature suggesting a ‘cochlear-type’ hearing loss occurring as part of neuromyelitis optica spectrum disorder. It raises the possibility of peripheral relapses of neuromyelitis optica spectrum disorder going unnoticed, resulting in patient morbidity, and highlights the importance of research within this area.

Background

Neuromyelitis optica spectrum disorder (NMOSD) is an inflammatory autoimmune central nervous system condition caused in the majority of cases by aquaporin-4 (AQP4) IgG antibodies (AQP4-Ab). Studies have shown that the regions of the cochlear and vestibular nuclei in the brainstem express AQP4,¹ and a handful of cases of retro-cochlear type hearing loss have been documented in the literature.^2–4 Additionally, one case with involvement of the extra-axial part of the vestibulocochlear nerve has been published in the literature.⁵ Interestingly, AQP4 has also been reported within the inner ear, both in epithelial cells within the organ of Corti as well as the cristae and maculae of the vestibular apparatus.⁶ This has led to the hypothesis that a ‘cochlear-type’ hearing loss and labyrinthine involvement could also occur in NMOSD.

Case presentation

A 54-year-old Caucasian woman diagnosed with AQP4-Ab NMOSD 9 years previously was referred to the specialist neuro-otology clinic with a 1-year history of right-sided tinnitus and low-to-mid-frequency moderate sensorineural hearing loss on pure tone audiometry (PTA) (figure 1A). Audiological testing shortly prior to the review in the neuro-otology clinic showed no change in her hearing thresholds on PTA for the right side, however new mild low-frequency left-sided hearing loss (figure 1B).

Figure 1.

(A) Pure tone audiogram (PTA) at onset of right-sided tinnitus, showing a moderate low and mid-frequency right-sided sensorineural hearing loss and normal left-sided thresholds other than at 8 kHz (mild dip to 25 dBHL). (B) PTA prior to assessment in neuro-otology clinic, showing new mild low frequency left-sided hearing loss. (C) PTA following steroid treatment, showing improvement in left-sided hearing thresholds back to baseline.

From the history there was nothing to suggest an infectious cause at the onset of the hearing loss, there were no cardiovascular risk factors and nothing to suggest Meniere’s disease. There was no history of episodic vertigo/dizziness or balance difficulty.

Examination revealed right beating nystagmus both on right gaze and following the head shaking test, indicating relative left-sided vestibular hypofunction. Otoscopy was normal.

Investigations

She had already been reviewed by otolaryngology 9 months previously who had arranged an MRI scan of the internal auditory meati which was normal.

Transient evoked otoacoustic emissions (TEOAEs) and distortion product otoacoustic emissions (DPOAEs), which both test outer hair cell function in the cochlea, were absent on the right (DPOAEs were only present at 6 and 8 kHz), suggesting cochlear dysfunction (figure 2A, B). TEOAEs and DPOAEs were present on the left, however it should be noted that emissions can still be present at a mild hearing loss level even if the cause of the hearing loss is cochlear dysfunction (figure 2C, D). The Auditory Brainstem Response test (ABR) showed normal waveforms bilaterally (at 80 dBnHL), suggesting no retro-cochlear pathology up to the level of the inferior colliculi.

Figure 2.

(A) Absent right-sided transient evoked otoacoustic emissions (TEOAEs) suggesting cochlea outer hair cell dysfunction. (B) Absent right-sided distortion product otoacoustic emissions (DPOAEs) other than at 6 and 8 kHz, suggesting outer hair cell dysfunction and in keeping with the low-to-mid-frequency sensorineural hearing loss shown on pure tone audiometry. (C) Present left-sided TEOAEs. (D) Present left-sided DPOAEs.

Vestibular testing showed normal cervical vestibular-evoked myogenic potential responses bilaterally (tested at 95 dBnHL), in keeping with normal bilateral function of the sacculocollic reflex pathway. Bithermal caloric testing showed a non-significant left canal paresis (9%) but a significant right beating directional preponderance (27%). There was normal central suppression of nystagmus. The directional preponderance on caloric testing, in combination with the clinical oculomotor findings, is suggestive of mild left-sided vestibular hypofunction.

Differential diagnosis

There was nothing in the history to suggest an infectious or ischaemic cause of her signs and symptoms, and although low-mid-frequency hearing loss is associated with Meniere’s disease, she did not present with episodic vertigo or aural symptoms and she did not meet the diagnostic criteria. She had already had a normal MRI scan and there was no evidence of central vestibular pathology on testing. The pattern of the hearing loss (low mid-frequency) starting on one side and then affecting the other (accompanied by mild vestibular hypofunction) is suggestive of possible autoimmune aetiology. Therefore, examination and testing were suggestive of peripheral dysfunction due to suspected autoimmune aetiology. We were suspicious of her underlying NMOSD being responsible as we had had another patient previously who had audiovestibular dysfunction, both peripheral and central, possibly related to her NMOSD.

Treatment

On advice of the neurology team, she was started on 5 days of oral methylprednisolone 500 mg, followed by a 2-month course of reducing oral prednisolone, starting at 40 mg daily and reducing by 5 mg every week. Proton pump inhibitor cover was given.

Outcome and follow-up

On completion of the steroid course, repeat PTA showed unchanged right-sided hearing thresholds, however the ‘newer’ left-sided mild low-frequency hearing loss had normalised (figure 1C). Gaze-evoked right beating nystagmus was still present. A further follow-up was arranged for video head impulse testing. This elicited a vestibulo-ocular reflex (VOR) gain of 0.93 on the right and 0.88 on the left horizontal planes. The values of the VOR gains and the asymmetry (5%) were within normal range.

A review of the patient’s AQP4-Ab titre levels revealed that at the onset of the right-sided hearing loss and tinnitus, her titre level was 200. Immediately prior to starting methylprednisolone (and following the onset of left-sided mild hearing loss), her level was 100. At the end of the prednisolone course, her level was 25. Therefore the titre level was eightfold greater at the time of the first attack, and fourfold greater after the second attack (but presteroid treatment), compared with after completing the steroid treatment.

Discussion

In this case, there is no evidence of central damage as indicated by the normal MRI and ABR testing, rather only labyrinthine damage as indicated by the abnormal PTA, absent right-sided otoacoustic emissions and evidence of mild vestibular hypofunction. There was nothing in the history to suggest an infectious or ischaemic cause of this damage, and the pattern of the hearing loss (low mid-frequency) starting on one side and then affecting the other is suggestive of an autoimmune aetiology. Furthermore, the improvement in the left-sided hearing thresholds in response to steroid treatment, as well as the reducing AQP4-Ab titre levels, would both support an inflammatory process as the aetiology. We propose this patient’s inner ear damage is due to autoimmune inflammation as part of her NMOSD. It may be that the right-sided damage had been present for too long before commencing the steroids for recovery to be possible on this side.

To our knowledge, this case is the first in the literature to suggest a ‘cochlear-type’ hearing loss occurring as part of NMOSD. Furthermore, it suggests the possibility that ‘peripheral’ relapses of NMOSD may go unnoticed and therefore untreated. It raises the question as to whether a low threshold for audiovestibular testing should be implemented and appropriate steroid treatment commenced to try to salvage/limit the damage caused to the labyrinth. Further work is needed in this area to evaluate whether peripheral audiovestibular damage is more common than currently thought in patients with NMOSD and whether treatment with steroids is beneficial in this group of patients with regard to their audiovestibular function.

Learning points.

• This is a case suggesting peripheral (labyrinthine) rather than central pathology, occurring as part of neuromyelitis optica spectrum disorder (NMOSD), resulting in audiovestibular dysfunction.

• ‘Peripheral’ relapses of NMOSD may go unnoticed and untreated and therefore a high index of suspicion is required to avoid patient morbidity.

• Further research is required in this field, specifically looking at the beneficial use of steroids in this cohort of patients.