Timothy C. Hain, MD • Page last modified: March 3, 2021
Goodall reviewed current understanding of AIED in 2015, and Strum reviewed treatment of AIED in 2020. Strum commented that AIED presently lacks an objective method of diagnosis as well as a consensus treatment protocol. We agree. This means that most of what we discuss here has the "garbage in: garbage out" problem -- we don't really know how to diagnose AIED, and we also don't have a consistent treatment strategy either.
The classic picture of autoimmune inner ear disease is reduction of hearing accompanied by tinnitus (ringing, hissing, roaring) which occurs over a few months. Variants are bilateral attacks of hearing loss and tinnitus which resemble Meniere's disease, and attacks of dizziness accompanied by abnormal blood tests for self-antibodies. About 50% of patients with AIED have imbalance. Because autoimmune inner ear disease usually affects hair cells in the inner ear, and because haircells are needed both for hearing and balance, generally there is both hearing loss and dizziness as symptoms. Less severe form forms have just hearing loss, because hearing is more vulnerable mainly because it has a more difficult job and does not have as much redundancy.
Nomenclature note: Some authors consider "AIED" to be a subtype of immune-mediated vestibular disorders, rather than a general category (Russo et al, 2019). We will not adopt their nosology, but will instead use the common system that AIED is a general term itself for immune-mediated inner ear disorders, that contains within it all other types of ear damage that can be attributed the immune system attacking the inner ear. It is difficult to imagine a place where this might not work, but perhaps if one injected in antibodies from another person that attacked the inner ear -- they would not be "autoimmune".
Subtypes of AIED
The immune system is complex and there are several ways that it can damage the inner ear. Both allergy and traditional "autoimmune disease" such as Ankylosing spondylitis, Behcet's, Systemic Lupus Erythematosis (SLE), Sjoegren's syndrome (dry eye syndrome), Cogan's disease, ulcerative colitis, Wegener's granulomatosis, relapsing polychondritis, rheumatoid arthritis, scleroderma (Amor-Dorado, 2008; Armin Deroee et al, 2009), Vogt-Koyanagi-Harada syndrome (Russo et al, 2019), and vitiligo (de Jong et al, 2017) can cause or be associated with AIED.
Allergy involving the inner ear is traditionally felt to be food related, but there is presently no agreement as to the importance of food allergy.
Ankylosing spondylitis (AS) is a progressive bone disease associated with fusion of the spine. Persons with severe cases of AS may be disabled because of their lack of flexibility. Although one might expect AS to be associated with conductive hearing loss, AS has been reported to be associated with a sensorineural hearing loss in about 28% of patients ( Alatas, Yazgan et al. 2005 )
Rheumatoid arthritis (RA) has higher prevalence. It is associated with hearing loss -- according to Jeong et al (2017), who stated "The prevalence of hearing impairment was higher in the subjects with RA than in those without RA, in both, the low/mid- and high-frequency categories (21.1% vs 7.5%, p < 0.001 and 43.3% vs. 26.2%, p < 0.001, respectively). In the multivariable logistic analysis, RA (odds ratios [OR] 1.47, 95% confidence interval [CI] 1.05-2.06, p = 0.025) was an independent risk factor of low/mid-frequency hearing impairment... " So in other words, RA is associated with "hearing impairment". This paper (which is available online) does not quantify the degree of hearing loss. It is our impression that it is rare that there is a substantial hearing loss in RA and that this study reporting high frequencies of mild impairment.
For Cogan's syndrome, see this page.
Susac syndrome (microangiopathy of ear, brain and eye) is a extremely rare disorder in which deafness, reduced vision, and encephalopathy may all present simultaneously. The encephalopathy manifests with headache, confusion, memory loss, behavioral changes, dysarthria and occasional mutism. The hearing loss is usually bilateral and accompanied by tinnitus and vertigo. According to Susac, it is generally a low-frequency sensorineural loss, resembling that of Meniere's disease. The reduced vision is caused by retinal artery occlusions. MRI images, which are invariably abnormal, reveal multifocal white matter lesions, including the corpus callosum (Susac et al, 2004). The lesions of the corpus callosum in Susacs are typically medial and centrally located (snow-balling) compared to lesions seen in MS that are more often seen on the undersurface of the corpus callosum at the septal interface (Susac, 2004). Treatment is with immunosuppresants (Clement et al, 2003).
Vogt-Koyanagi-Harada syndrome (VKH), is another autoimmune disorder combining uveitis, vitiligo (patches of depigmented skin), poliosis (white eyebrows), and sometimes inner ear symptoms. It is mainly recognized by the uveitis, and has some overlap with Cogan's syndrome (which also has eye inflammation), and Susac (which has eye findings as well). VKH has some overlap with sympathetic ophthalmia (another autoimmune eye disease, triggered by eye trauma). If one sees the symptom of poliosis accompanyng uveitis and inenr ear symptoms, , one should think VKH.
A relatively new autoimmune inner ear disease is "IgG4". According to Li et al (2017), "IgG4-related disease (IgG4-RD) is an idiopathic inflammatory condition that causes pseudotumor formation in single or multiple organs, including those of the head and neck. Temporal bone involvement is rare, with only 3 cases of unilateral temporal bone IgG4-RD described in the literature." This disease is a biopsy diagnosis. This disease could be more common than we know, as it requires a biopsy of the area of the temporal bone. Perhaps some of those cases of mastoiditis that the literature says to ignore are actually IgG4 disease.
Not all individuals with bilateral sensorineural hearing loss have autoimmune disease. Genetic defects, infections, toxins, advanced age, noise exposure, and conditions of mysterious origin all account for some cases.
AIED is rare, probably accounting for less than 1% of all cases of hearing impairment or dizziness. The precise incidence is not settled. Das et al (2019) offered the estimate of 5 cases/100,000 population. Russo et al (2019) suggested that it accounts for about 1% of all inner ear disease. About 16% of persons with bilateral Meniere's disease, and 6% of persons with Meniere's disease of any variety may be due to immune dysfunction. As Meniere's has a prevalance of about 2/1000, 6% of them would be 12/100,000. We would not expect a large number of patients outside of Meniere's, as Meniere's is by far the most common cause of these symptoms.
In our clinic database in Chicago at Chicago Dizziness and Hearing, we have only 19 patients with the diagnosis of AIED, out of a total number of patients of roughly 25,000. Thus AIED accounts for only 19/25,000 or less than 1/1000 of the patients that we have evaluated (mainly with dizziness). Our 25,000 patients include many without any hearing issue as many are just dizzy. To try to put this into context, we have about 990 patients with Meniere's in our database. Thus our clinic figures would suggest a ratio of about 20/1000 AIED to Meniere's, far less than the number of 6% -- 60/1000 figure offered above. As the prevalence of Meniere's is known to be about 1/2000, accepting our alternative method of calculation would imply that the prevalence of AIED should be about 19/990*1/2000, which is very low, roughly 1/100K. A disease is defined as "rare" in the USA, when it affects fewer than 200,000 people. As the US has about 325 million people, this would mean that only about 3118 people have AIED in the US, and it is "rare". Taking the higher estimate, 12/100K, it would still be rare. Of course, these figures are extremely rough, depend on quite a few assumptions about the distribution of patients in the clinic -- i.e. the # of patients with Meniere's. Rather large differences in this estimate might also be produced depending on how tightly one defines AIED. We would far prefer to see a prospective population based study.
What Causes Autoimmune Inner Ear Disease ?
The cause of AIED is generally assumed to be related to either antibodies or immune cells that cause damage to the inner ear. There are several theories as to how these might arise, analogously to other putatative autoimmune disorders:
Bystander damage: In this theory damage to the inner ear causes cytokines (or other autoimmune "bad actors") to be released which provoke, after a delay, additional immune reactions. This theory might explain the attack/remission cycle of disorders such as Meniere's disease. There are certainly cytokines in the cochlea including interleukin-1A, TNF-alpha, NFkB P65 and P50, and IkBa (Adams, 2002). Other mediators may be ICAM-1, IL-1beta, IFN-gamma, and IL-17(Russo et al, 2019). Of course there are cytokines everywhere so this doesn't really mean much other than immune responses are associated with inflammation which releases these cytokines.
Drugs that block TNF such as etanercept (see treatment section) seem to be potentially effective in AIED (Rahmen et al, 2001). The endolymphatic sac lumen expresses TNF-alpha (Satoh et al, 2003), which may be another way wherebye Meniere's disease is linked to AIED -- or maybe not as this information doesn't tell us how important this expression is. Other autoimmune disorders such as Crohn's disease also seem to be linked to TNF, and other ear diseases such as otosclerosis also are linked to TNF-alpha. (Karosi, Konya et al. 2005)
Cross-reactions: In this theory, antibodies or rogue T-cells cause inner ear damage because the ear shares common antigens with a potentially harmful substance, virus or bacteria that the body is fighting off. This is presently the favored theory of AIED. COCH5B2 has been reported to be a target antigen in AIED (Boulassel et al, 2001).
Intolerance: The ear, like the eye may be only an partially "immune privileged" locus, meaning that the body may not know about all of the inner ear antigens, and when they are released (perhaps following surgery or an infection), the body may wrongly mount an attack on the "foreign" antigen.
Occasionally there is deafness in one ear, following trauma or surgery performed on the opposite ear. A longer discussion of "sympathetic inner ear disease" is found here.
An related question to "what causes AIED", is what damage is done by AIED ? One might reasonably imagine damage to hair cells, damage to the auditory nerve or ganglion, or structural changes. Lobo et al (2018) recently suggested that hydrops was found in 11/12 patients with AIED. Hydrops is a concept developed in the context of Meniere's disease, which overlaps to some extent with AIED. It really makes no sense at all that a plumbing problem (hydrops) should be caused by immune problems -- this is like suggesting that sinks are backing up due to termites.
There is evidence that genetically controlled aspects of the immune system may increase or otherwise be associated with increased susceptibility to common hearing disorders such as Menieres disease. Bernstein and associates reported that 44% of patients with Menieres, otosclerosis and striatal presbyacusis had one particular extended MHC haplotype (Dqw2-Dr3-c4Bsf-C4A0-G11: 15-Bf:0.4-C2a-HSP70:7.5-TNF), compared to only 7% of controls. Sudden hearing loss in Koreans that does not recover is also associated with HLA-DRB1*04, DQA1 03 and 05 (Yeo et al, 1999; Yeo et al, 2001). The author has also found an association (in the US) with certain types of HLA-types and variants of vertigo in caucasians (unpublished). On the other hand, a recent study by Lopez-Escamez and others performed in Spain found no difference in HLA antigens between 54 patients with definate MD and 534 normal controls (Lopez Escamez et al, 2002). The genetic background of HLA studies is important and it is possible that one group might find HLA differences which are not found in another.
These data are thus conflicted. If there is indeed an association with HLA, at least in certain populations, it would suggest that more of Menieres disease and other progressive syndromes may be caused by immune dysfunction than is presently generally thought. It is important to remember that HLA-typing is relevant when considered in the context of the patient's genetic background. In other words, studies of Korean subjects for example, such as reported by Yeo, may not apply to persons of non-Korean ethnicity.
The diagnosis is based on history, findings on physical examination, blood tests, the results of hearing and vestibular tests, MRI scans, and response to immunosuppressive medications. The usual clinical picture is a subacute bilateral progressive sensorineural hearing loss.
|Patient with a hearing loss that partially responded to steroids over several weeks.|
Steroid responsiveness is the most useful method of making the diagnosis, and ordinarily the diagnosis is made by observing a bilateral progressive sensorineural hearing loss that responds to steroids. This is a rather difficult criterion to meet, as it requires three features -- bilateral, progression, and steroid responsiveness. There might be few people where steroid responsiveness has been established, as this generally requires contact with a subspecialty physician. The observation of progression requires a time-series of audiograms, quite different from the "point of service" type audiograms used in most settings.
Other tests may be proposed based on the clinical situation.
As auditory neuropathy can present with a progressive bilateral sensorineural hearing loss, ABR testing should be done in persons with enough hearing for the test to be practical. Otoacoustic emmission tests should be done in those in whom ABR testing cannot be done. MRI scans ot the brain are useful to diagnose Susac's syndrome (see above), as well as to exclude possible confounding disorders, such as acoustic neuroma.
While specific tests for autoimmunity to the inner ear would be desirable, at this witing (2020) we know of none that are both commercially available and proven to be useful. (Garcia Berrocal et al, 2002).
Having an autoimmune disease, per se, combined with hearing loss is not a good way of establishing that the hearing loss is autoimmune, as both autoimmune disease and hearing loss are very common. In other words, coincidence is not the same as cause. Here is a partial list of blood tests.
A small study suggested that FDG PET scans may be useful in AIED. (Mazlumzadeh et al, 2003). More investigation of this modality is needed before it's role in diagnosis can be defined. The cost is clearly prohibitive.
As there are no specific tests for AIED, a common approach is to look for other evidence for autoimmune involvement.
Blood tests for autoimmune disorders, ordered from most to least useful, include:
As commented above, positives on these blood tests do not prove that hearing loss is caused by autoimmune disease -- some of them simply document inflammation. Response of hearing to steroids is far more specific. It is best to get these tests done prior to starting immunosuppressive drugs.
Blood tests for conditions that resemble autoimmune disorders, again from most to least useful, include:
Recently it has been suggested that blood levels of TNF, tumor necrosis factor, is both diagnostic and predictive of treatment response (Svarkic, 2012). As TNF is a nonspecific inflammatory cytokine, we are dubious as we think that TNF should logically be elevated in many conditions.
MRI testing is mainly done to exclude other entities, but occasionally enhancement is seen of the cochlea.
For the most part, this is the differential diagnosis of progressive bilateral sensorineural hearing loss. (Kishimoto et al, 2013)
As an overview, there is no evidence that anything really "works". Strum et al (2020), wrote in a meta-analysis of treatment for AIED: "The high quality randomized controlled trials included did not show benefit of any other medication in maintaining the initial steroid burst improvement in hearing. " If we keep this in mind -- the cost/benefit ratio for all medical treatments is very poor. This is the reason that some clinicians advise just letting the disease run its course and follow up with a cochlear implant. This logic is a little harder to sustain in the unfortunate people who lose both hearing and vestibular function, as so far, there is no successful vestibular implant.
Usually treatment of AIED starts with a "shotgun" type treatment - -a drug that suppresses everything. These are “shotgun” type drugs that suppress the entire immune system. McCabe stated that Cytoxin was the “cornerstone” (McCabe, 1989). However, of course these drugs have severe side effects. Steroids is another "shotgun" type drug.
As an example, in cases with a classic rapidly progressive bilateral hearing impairment, a trial of steroids (prednisone or dexamethasone) for 4 weeks may be tried. High dose steroids are not used for long periods of time because of systemic side effects. Thus, in persons with response to steroids, in most cases a "steroid sparing" drug -- in the past a chemotherapy type of medication such as Cytoxan (cyclophosphamide) was used over the long term (Sismanis et al , 1994; Sismanis et al, 1997). While often patients are converted over to methotrexate, this was shown to be ineffective by Harris (2003). Currently patients are often being treated with "Biologics", which have relatively less toxicity than chemotherapy drugs.
Local use of steroids seems more sensible -- i.e. with injections through the inner ear. Steroids administered this way have the main problem that they rapidly leave the inner ear. Attempts have been made to develop slowly dissolving forms, such as the "OTO-104" trial (Lambert, 2012). We are presently dubious that this is a good idea. More discussion of steroid treatment for inner ear disease is found under the category of SHL (sudden hearing loss) as well as Meniere's disease..
Slightly narrower treatments are those that only target humoral immunity:
Plasmapheresis may be beneficial (Luetje and Berliner, 1997; Hussain et al, 2005). Plasmapheresis requires periodic visits using a machine similar to a dialysis unit. We do not think this is a good idea.
Intravenous immunoglobulin has been reported successful in a single case with scleroderma (Deroee et al, 2009). This is just one case.
We have one patient with RA (rheumatoid arthritis) and AIED who is doing very well on Orencia ( Abatacept). This is a drug approved for rheumatoid arthritis, that prevents T-cell activation. We are not sure if it is a generally good drug for AIED, and as it is expensive and difficult to prescribe due to insurance company restrictions, we may never know. This type of drug would seem a good choice for autoimmune disorders mediated by cellular immunity.
TNF-α is a cytokine -- a small molecule that turns on immune responses. Interfering with TNF-α is a succcessful treatment for several common autoimmune disorders. Until just recently, these drugs were given off-label (translation = extremely expensive) through injections or infusions. Recently however, there has been an effort to administer these drugs directly into the ear. We will start with this dissussion.
Intratympanic treatments using anti-TNF agents. A recent exciting development in treatment of AIED is the use of local TNF-α agents injected through the ear drum. This reduces the opportunity for toxicity.
Derebery et al (2014) in an early study of golinumab (one brand name is Simponi) wrote that "The TNF-alpha inhibitor GLM stabilized hearing in 3 of 7 per-protocol subjects with AIED and allowed a complete tapering off of prednisone in those 7 subjects. Studies with larger samples sizes are warranted."
Mata-Castro et al (2020) recently wrote about infliximab (Remicade) saying : "Infliximab intratympanic infiltration improves the hearing threshold in patients with immune-mediated hearing loss. The effect of improving the hearing threshold is higher in low frequencies and persists within 3 months of the infiltration. The administration of intratympanic infliximab is an effective and safe technique."
Overall, this appears to be promising.
Oral or injectable anti-TNF drugs:: Etanercept (Enbrel) is a promising agent for treatment of AIED (Rahmen et al, 2001; Wang et al, 2003). Enbrel is an anti-TNF (tumor necrosis factor) drug. TNF is an inflammatory cytokine (see above). Wang et al recently reported that etanercept given acutely in sterile experimental labyrinthitis resulted in much better hearing results in an animal model. On the other hand, Cohen et al (2005) reported that etanercept was no better than placebo in persons with chronic AIED in restoring hearing lost to the disease. We don't feel that this is surprising or very meaningful, as we think it is generally unreasonable to think that any drug, including etanercept, could cause hair cells to come back to life from the dead. In our clinical practice, we have had generally very good results in our patients using anti-TNF drugs with steroid responsive, progressing AIED and we feel that because of the study design of the Cohen paper (chronic hearing loss, no requirement for steroid response), that etanercept is not completely ruled out. Etanercept is given as an subcutaneous injection once/week, usually between 25 and 50 units. Etanercept has generally been well tolerated but according to the manufacturer's information, people on etanercept have developed serious infections (2%), nervous system disorders, and depression/personality disorders (1%). A recent large study suggested that there was increased risk (which is normally quite low) of various brain inflammatory events (Kunchok et al, 2020). Or in other words, anti-TNF drugs might make very unlikely events more likely.
A related agent, (infliximab) Remicade, was not found useful for AIED, but this study was based on only a handful of cases (Pyykko et al, 2002). Inflixamab was found useful in another study (Heywood, 2013). Inflixamab is also not suitable for home use. There are newer agents that are in the drug pipeline that will need to be tested for their efficacy. Of the newer anti-TNF drugs, the most interesting is Humira (adalimumab), which is another anti-TNF drug. This drug is probably equal to etanercept.
Drugs already available in the world that are also anti-TNF agents include thalidomide, pentoxifylline (a vasodilator used for poor circulation), and rolipram (an antidepressant available in Japan and Europe). These drugs have not been tried in AIED. Thalidomide is extremely unsafe, but pentoxifylline might be worth investigating.
None of these drugs has an official FDA indication for AIED. There is justifiable concern about these drugs affecting other health problems such as how well the body fights infection or kills tumor cells. In controlled studies of all TNF-alpha blocking drugs, more cases of lymphoma have been noted in treated patients than controls. Lymphomas are also often seen with use of other immunosuppresants including azathiprine and/or mercaptopurine. It is also generally felt that when these drugs are in use there should be increased vigilance for reactivation of tuberculosis.
Another significant problem with use of etanercept as well as related biologic category drugs is it's very high cost and lack of uniform insurance coverage. This is an expensive drug, and insurance companies often simply balk at paying for it, saying that it is experimental. If one has an autoimmune disease where etanercept is already approved, there generally is no problem. However, you may be out of luck if you don't have an "indicated" disease such as rheumatoid arthritis or psoriasis.
While new and expensive immune suppressant drugs are in abundant supply, for the most part, nobody knows whether or not they work in AIED. Or more precisely, strong randomized trials are missing as of the end of 2019 showing that these drugs work for AIED (Strum et al, 2020). Generally speaking, it seems likely that drugs that work for psoriasis, rheumatoid arthritis, and ulcerative colitis may also be helpful for AIED. Also from general principles, "rifle" like drugs such as monoclonal antibodies, are to be prefered to "shotgun" type drugs such as steroids.
For example, Stelara (Ustekinumab) -- a monoclonal antibody drug -- is also used for psoriasis, but rather than being an anti-TNF agent, it is directed against interleukin 12 and interleukin 23. Stelara, like the anti-TNF agents, increases the risk of cancer, depresses immune responses, and reduces one's ability to fight TB. Stelara needs to be taken far less frequently than does etanercept. It is unknown whether ustekinumab is effective in AIED. We will be unlikely to know anytime soon as it is prohibitively expensive.
There have been a few promising studies of IL-1 receptor blockers (Goodall & Siddiq, 2015) such as Anikinra (Vambutas et al., 2014). According to Brant, it is difficult to evaluate the trials to date. (Brant, Eliades, & Ruckenstein, 2015) .
Cochlear implants are useful in AIED as a last resort (Wang et al, 2010). Cochlear implants are extremely expensive, but there are no major issues with insurance coverage once one becomes "unaidable". In our opinion, they are very reasonable alternatives to immune suppressants as immune suppressants may increase the risk of serious infection and cancer. This logic may be changed if the new approach using intratympanic TNF-a inhibitors is borne out.
In animals, attempts have been made to treat variants of AIED with oral collagen (Kim et al, 2001). Relapsing polychondritis is a disorder in which there may be antibodies to collagen and acquired deafness.
Autoimmune inner ear disease is very rare, making it difficult to study. One can speculate that there might be effective treatments that simply have not been discovered. For example, there are numerous potential treatments that have not been tried in a formal way. Gamma globulin infusions, given monthly, are useful in numerous autoimmune disorders. This treatment is very expensive, which limits its use. Immune modulating drugs such as are used for treatment of MS (beta-interferon, alpha-inteferon, copaxone) have not been tried in AIED, to this author's knowledge. Other medications that have coincidental suppression of immune responses, such as minocycline, or other anti-TNF drugs (see above), might be tried.