Discussion
AIED is a rare cause of bilateral SNHL, accounting for <1% of cases, with a yearly incidence of <5 per 100 000.[2] It was described by McCabe in 1979 following a case series involving 18 patients who presented with bilateral SNHL (>30 dB for at least 3 frequencies) that showed clinical and audiometric improvement to immunosuppressive therapy, with no clear cause. [1, 4] The hearing loss in AIED is characterised by its bilateral, asymmetric and fluctuating nature.[6,7] Patients can also report tinnitus and ear fullness (25-50%), and vestibular symptoms such as disequilibrium, ataxia, nystagmus and episodic or positional vertigo (50%).[2, 7] It has a rapidly progressive onset, developing between 3 to 90 days.
The pathogenesis is not completely understood and studies on AIED have been difficult due to (1) the limited anatomical access to the cochlea, (2) unreliable data from peripheral blood immune system and (3) a lack of an ideal animal model. [5] The proposed mechanism for AIED is thought to be an uncontrolled, combined humoral and cell-mediated reaction against inner ear antigens[2, 3] which promotes autoantibody formation and pro-inflammatory T-cell responses. This results in: (1) cochlear vascular injury due to immune complex deposition, (2) autoantibody-related microthrombosis and (3) vascular changes involving electrochemical disturbances and impaired neurosignalling.[2]
When the inner ear is the only organ affected, AIED is considered ‘primary’. If it occurs with systemic autoimmune diseases such as systemic lupus erythematosus, rheumatoid arthritis and Sjogren’s syndrome, it is considered ‘secondary’. Secondary AIED accounts for up to 30% of cases. [2] The Harris AIED classification [7] further categorises the condition into 6 different types. Our patient’s condition co-exists with UC and is classified as secondary AIED, or ‘type 2: rapidly progressive bilateral SNHL with systemic autoimmune disease.’[7]
There is a lack of standardised diagnostic criteria and pathognomonic tests for AIED so diagnosis can be challenging and is often based on exclusion. Diagnosis typically relies on 3 factors: clinical evaluation, blood tests and response to corticosteroids. [2, 7] Laboratory tests must be conducted in all patients with suspected AIED. This includes routine blood tests and autoimmune screening (Figure 1) to investigate for underlying systemic autoimmune disease and exclude other differential diagnoses. [2] There are no confirmed serological markers, but a few have been proposed such as heat shock protein 70 (HSP70) antibody, an autoantibody that targets an inner ear antigen. [8] It was thought to be a predictor of steroid responsiveness but in a further study, it was found to be insufficiently sensitive to detect AIED and lacked specificity to predict steroid responsiveness. [5, 7] A response to corticosteroids is defined as meeting these 2 requirements: (1) improvement in SNHL >15 dB at any frequency OR >10 dB in at least 2 consecutive frequencies OR an increase in pure-tone air conduction threshold loss at any frequency or decrease in WRS. [2, 7] MRI also plays a role in the diagnosis to rule out retrocochlear pathologies. [2]
The clinical features of AIED can mimic other inner ear pathologies such as sudden sensorineural hearing loss (SSHL) and Meniere’s disease (MD),[2, 8] and may often be misdiagnosed. It can be differentiated by observing the timing of disease onset; for example, hearing loss in SSHL is characterised by an acute onset of <3 days and in MD, this typically occurs over years.[2] A case report by Ho et al. [2016] discussed a patient with AIED presenting as MD [8]who, after 4 years of symptom onset, presented with psoriasis, joint pains and anterior uveitis. It was then that a diagnosis of AIED was suspected. The patient was started on a trial of corticosteroids, which he responded to, and was switched to steroid-sparing therapy.[9] In our patient’s case, he was initially diagnosed with bilateral labyrinthitis following his unremarkable MRI and blood test results. The diagnosis of AIED was considered approximately 5 weeks after his initial presentation after he re-presented multiple times with a deterioration in his hearing shortly after finishing his course of steroids. However, it took almost 1 year from his initial presentation and referral to an ENT specialist that undertook cortical evoked response audiometry to definitively confirm the diagnosis of AIED.
AIED is one of the few forms of SNHL which may be reversible so prompt medical management is crucial as it may improve the rates of reversing hearing loss. [4] The patients in McCabe’s study in 1979 were initially treated with cyclophosphamide and corticosteroids[1] however due to the substantial risks associated with long-term cyclophosphamide use, including malignancy and myelosuppression, most clinicians favoured high-dose prednisolone (60mg OD) as the primary treatment for AIED. [6]High-dose steroid use is also associated with health risks and adverse effects [6] such as hyperglycaemia, weight gain and fractures secondary to osteoporosis. A study found that with good patient education and monitoring, it is a safe and effective treatment for AIED [10] but is not effective for long-lasting management and is associated with episodes of relapse.[11]
Although corticosteroids remain the mainstay treatment, there are varying levels of steroid-responsiveness initially and throughout the treatment. One study evaluating the efficacy of AIED treatment reported that the overall patient response rate to oral prednisolone was 69.7%.[6, 7] Another study found that only 14% remain corticosteroid-responsive after 34 months, most notably in those who present with repetitive deterioration in hearing requiring corticosteroid treatment. [6] Our patient received numerous courses of steroids during this treatment process and initially saw an improvement in his symptoms with prednisolone 40mg OD later required 60mg OD to get the same response. He would also experience a worsening of his symptoms when he is on lower doses or off steroids, relapsing as early as 3 days after stopping treatment.
There is a risk of further decline in hearing in the absence of therapy and yet, there is no standard treatment protocol for AIED following corticosteroids as randomised controlled trials (RCTs) are limited.[7] Studies regarding alternative steroid-sparing therapies have yielded variable results, with much of the data relying on case reports and case series. [3, 11]Alternative therapies that have been trialled include methotrexate, azathioprine and immunomodulators [11] such as biologics, including TNFα inhibitors (etanercept and infliximab), CD20 inhibitors (rituximab) and IL-1 inhibitors (anakinra and canakinumab). If medical treatment fails and hearing is lost, cochlear implantation is an effective treatment option for these patients.[4]
The efficacy of methotrexate was assessed in a multicentre clinical trial as a potential steroid sparing therapy in corticosteroid-responsive patients but did not demonstrate a greater effect than the placebo. [5] A case report described a presentation of relapsing AIED in a 35-year-old woman who had a significant response to methotrexate and azathioprine dual therapy. This combination yielded positive outcomes and was well-tolerated. [12] Our patient was initially tried on methotrexate in combination with azathioprine which he was already taking for his UC but unfortunately showed little to no improvement.
Our patient’s AIED is classified as secondary and more specifically, it fits the Harris AIED classification type 2 that often worsens with flare of the autoimmune condition. This type is steroid-responsive and can be managed with targeted therapies that are effective against their underlying condition. [7] The AIED specialist centre and gastroenterology agreed that AIED is a rare manifestation or complication of UC. This, combined with the failure of achieving remission for his UC despite being on azathioprine and receiving multiple courses of high-dose steroids, led the MDT to commence him on infliximab. The patient showed clinical and audiometric improvement after two loading doses of infliximab infusions. His hearing remained stable, and he also reported improvement in his bowel habits and symptoms of UC since starting therapy. The patient has tolerated infliximab well and has remained on the infusions.
On literature review, there are two similar case reports to our patient’s presentation. One reported a case of a 49-year-old man with AIED who was responsive to corticosteroids but experienced frequent relapses and progressive deterioration of hearing. He was tried on multiple agents, but these failed to stabilise his symptoms. After receiving infliximab treatment, he was noted to have a sustained improve in his hearing and tinnitus. There was an attempt to discontinue his treatment after 46 weeks, but this led to a rapid relapse of his condition, though his hearing quickly recovered after restarting infliximab. [13] Another case report presented a case of AIED in a patient with Crohn’s disease that showed improvement to anti-TNFα therapy, halting the progression of hearing loss, as well as improving hearing by an average of 15dB across all frequencies. The patient’s hearing remained stable after. [14]
A pilot placebo-controlled study on anti-TNFα therapy, etanercept, in AIED however resulted in a different outcome. [15]This involved 20 patients with AIED in a 12-week blinded placebo-controlled RCT of subcutaneous etanercept 25mg twice weekly. The efficacy of etanercept therapy failed to exceed the placebo response. Another study on 8 patients with steroid-refractory AIED concluded that systemic infliximab therapy did not improve hearing loss.[16]
Although studies on biologic therapy have variable results on hearing improvement, they have shown benefits on treating other aural symptoms such as ear fullness, vertigo and tinnitus, [8]with a retrospective study reporting that >80% of patients demonstrated improvement in these symptoms on treatment with adalimumab and rituximab. [7] Some studies involving methotrexate have also demonstrated a similar result, with one reporting an 80-100% improvement rate in vestibular symptoms.[9] Steroid-sparing therapy should therefore not be eliminated based on lack of improvement in hearing alone.
The disparity in response could potentially be explained by factors such as timing of the treatment relative to corticosteroid use, the type of anti-TNFα therapy used, and route of administration.[5] The later adaptive immune responses in AIED are often influenced by the early expression of cytokines during the innate immune response, such as TNF. Elevated levels of TNF are identified to be predictive of corticosteroid-responsiveness in AIED. Experimental observations reveal that peripheral blood immune cells from corticosteroid-responsive patients release high levels of TNF in vitro and these are reduced significantly on administration of dexamethasone. It is possible that the initial use of corticosteroids prior to anti-TNFα administration causes an excessive reduction of the TNF levels, potentially compromising its effectiveness in placebo-controlled trials. [5]