Timothy C. Hain, MD. • Page last modified: March 5, 2021
Sudden hearing loss (SHL) is defined as greater than 30 dB hearing reduction, over at least three contiguous frequencies, occurring over 72 hours or less. It occurs most frequently in the 30 to 60 year age group and affects males and females equally. Although called sudden, it seems unlikely that hearing loss is abrupt but rather it probably evolves over a few hours.
SHL can affect different people very differently. SHL is usually unilateral (that is, it affects only one ear); and is often accompanied by tinnitus, aural fullness, vertigo, or any combination. In fact, 90% have tinnitus with SHL. The amount of hearing loss may vary from mild to severe, and may involve different parts of the hearing frequency range. SHL may be temporary or permanent. About one third of people with SHL awaken in the morning with a hearing loss. A discussion of sudden bilateral hearing loss is found here.
SHL also should be "sensorineural" meaning that it is not caused by a mechanical process such as ear wax (which causes a "conductive" loss). Our experience is that most patients who complain of "sudden hearing loss" actually have ear wax impactions. Looking in the ear canal is the usual way these people are quickly identified. Sometimes the ear wax is deep inside. These people are not "diagnosed" as SHL, although they do have sudden -- hearing -- loss.
Another usual exclusion from SHL is bilateral hearing loss. It is very rare that hearing loss occurs in both ears within 72 hours, so generally the timing definition of SHL alone leaves this situation out. However, sometimes it is difficult to know -- such as someone with a preexisting unilateral hearing loss, who loses hearing in the other ear. Kishimoto et al (2014) reported on the clinical features of "rapidly progressive bilateral sensorineural hearing loss". They defined "rapid" as hearing deterioration progressing over a year in both ears. Of course, most of the processes that they noticed were either disorders that affected spinal fluid (such as meningitis or superficial siderosis), the body as a whole (a variety of autoimmune inner ear disease), and bilateral forms of unilateral ear disorders. Oddly, in this study they excluded bilateral Meniere's disease.
SHL seems to differ in characteristics according to whether it is primarily low-tone hearing loss (as occurs in early Meniere's disease), or having other patterns (Yoshida et al, 2017). This suggests that there may be two different groups and perhaps should be two different treatment approaches as well.
Sudden hearing loss is associated with vertigo in between 20 to 60% of patients (Rambold et al, 2005). The wide spread probably depends on what you define as vertigo -- a little bit of dizziness, or a profound sensation of rotation. Patients with vertigo also usually have worse hearing (Niu et al, 2015). This makes sense -- if you are have more symptoms, you are sicker.
There is presently no convincing evidence that any oral or intravenous treatment for idiopathic SHL is better than placebo. The question remains open regarding injections of steroids through the ear drum.
It is estimated that SHL occurs in 27 person/100,000 (Alexander and Harris, 2013), with the incidence increasing with age -- 11/100K for persons < 18 years old, to 77/100K for persons over 65. However, statistics like these are difficult to confirm --- SHL is thought to be often ignored, and also it recovers spontaneously as well. Our guess is that SHL is more common than this.
According to Zhang et al (2015), in China, in a study of about 250 people, SHL is commonly seen in middle aged people. Half of their study population had hyperlipidemia. When CT or MRI's were available, 14.4% were abnormal.
SHL is often associated with preexisting hearing loss (Koo et al, 2015). Thus people who already have ear damage -- are more likely to have more hearing loss documented going forward. And probably people who care enough about their hearing to get it tested in the past, are more likely to keep caring about it when they have a change.
In our medical practice in Chicago, of the 54 SHL patients in our database as of 2015, the median age was 60. Of course, statistics like this do not reflect the incidence in the population, as the composition of the underlying population is certainly not evenly distributed across age. In other words, there is probably a combination of increasing incidence with age combined with the age composition of the Chicago population from which this graph was drawn.
The short answer is that generally nobody knows.
|Autoimmune||Vascular||Neurological Neoplastic||Trauma or Toxin||Infections and viral|
|Autoimmune inner ear disease (AIED)||Cardiopulmonary bypass||Acoustic Neuroma||Large vestibular aqueduct syndrome||Cryptococcal Meningitis|
|Cogan's syndrome||Red blood cell deformability||Contralateral deafness after acoustic neuroma surgery||Inner ear concussion||Cytomegalovirus|
|Lupus||Sickle cell||Focal pontine ischemia||Inner ear decompression sickness||Herpes-simplex I|
|Meniere's||Small vessel disease||Leukemia||Otologic surgery||HIV|
|Polyarteritis nodosa||Vascular disease associated with mitochondriopathy||Meningeal carcinomatosis||Ototoxicity||Lassa Fever|
|Relapsing polychondritis||Vertebrobasilar insufficiency||Migraine||Perilymph Fistula||Meningococcal meningitis|
|Ulcerative Colitis||Blood dyscrasias||Multiple sclerosis||Temporal bone fracture||Mumps|
|Wegeners's granulomatosis||Intralabyrinthine hemorrhage||Myeloma||CSF leak, such as caused by lumbar puncture||Rubeola, Rubella, syphilis, Toxoplasmosis|
Table adapted from Wynne, 2003
Although some hold that this disease is generally idiopathic (of unknown cause), the differential diagnosis includes viral disease, Lyme disease and its relatives (Lorenzi et al, 2003), vascular disease (1%), autoimmune phenomena, perilymph fistulae and Meniere's disease, and acoustic neuroma (about 4 to 6% of SHL -- see table above and Daniels et al, 2000 for a longer list of diagnoses). Of course, if one knows what causes SHL, it isn't idiopathic (by definition).
Viral disease has been claimed to be the basis for about 60% of all cases of SHL. Viruses detected at a study at the Massachusetts Eye/Ear infirmary included influenza type B, CMV (Seguira et al, 2003), mumps, rubeola, and varicella-zoster (Harris, 1998). Others include measles, herpes-1, hepatitis and infectious mononucleosis. Many of these are in the herpes family.There are sporadic reports of other organisms as well (e.g. ehrlichiosis, Bhalia et al, 2017). Immunization is not associated with SHL (Baxter et al, 2016). One would expect that a viral "SHL" would progress over hours.
A temporal bone study of 17 bones from the Mass Eye/Ear infirmary suggests that pathology does not support the concept of membrane breaks (e.g. Meniere's), perilymphatic fistulae or vascular occlusion (Merchant et al, 2005). An additional study of 11 more temporal bones in 2013 (Linthicum et al), was suggestive of a viral etiology. In our view, these studies included too few temporal bone samples to be relied upon. More temporal bone donations are needed. Interestingly, the laboratory that reported the recent 11 bones, had been in operation for 50 years. This tiny number of samples over 50 years points out that the infrastructure to gather temporal bones in the USA is suboptimal. We ourselves have had situations where a patient indicated a willingness to have their temporal bone "harvested" but a "harvester" -- an otologic surgeon -- was not available. Once this happens, one loses one's enthusiasm in encouraging patients to be generous.
Vascular: Some authors maintain that vascular disease is the most likely cause (Rambold et al, 2005). This is largely a conclusion based on exclusion of other causes, although Umesawa et al (2017) reported that cardiovascular risk factors including smoking and diabetes were significantly more common in men in SHL. To us, this doesn't seem like very important as there are just too many qualifiers here. Also supporting the idea that vascular causes are important, is the observation that SHL following surgery outside the inner ear, is most commonly associated with cardiopulmonary bypass surgery (Page and Peters, 2015). One would expect that a vascular hearing loss would occur within seconds, or perhaps have some stuttering on/off hearing before it went "out". Because the cochlea (hearing part of ear) is very sensitive to brief ischemia, it would seem plausible that one could have a temporary vasospasm of the labyrinthine artery, and end up with hearing loss but little vestibular damage.
In general, when viruses or vascular etiologies are held out as the "cause" of a particular illness, it often means that the details are unclear. Recently there has been some weak recent evidence that blood flow plays a role as some measures of hearing improve in animals after a "stellate ganglion block" that improves blood flow (Firat et al, 2008). High levels of cholesterol and low levels of Coenzyme Q are associated with SHL (Cadoni et al, 2007). Use of some agents that affect blood flow -- largely increase it -- , such as sildenafil (Viagra) and Tidalafil (Cialis) but not vardenafil (Levitra) have been associated with hearing loss (McGwin, 2010; Khan et al, 2011; Wester et al, 2018). Fortunately these are very rare occurences -- perhaps just due to chance. Against this general idea is the observation that factor V Leiden (a common cause of increased blood coagulation) does not increase the chance of SHL (Shu et al, 2015). Also, the previously mentioned study of Merchant et al in temporal bones (2005) is against the vascular hypothesis.
In addition to ischemia (the inner ear has one main artery), there is also the possibility of hemorrhage. Just a few cases of this have been reported. Kim et al (2017) reported that about 10% of MRI's of 60 patients with SHL were suggestive of bleeding. Similarly, Wu et al (2018) reported presumed intralabyrinthine hemorrhage in 5/25 pediatric patients with SHL, and Chen et al (2018) reported 42 ptients. We are a little dubious about this, as we don't see why there should be bleeding in the inner ear, but not elsewhere. In addition, we are not entirely convinced that every abnormal MRI with bright signal necessarily means that there was a bleed. On the other hand, there are reports of patients with coagulation disorders having inner ear bleeds.(Cherchi et al., 2006). It is our thought that these patients should logically have both hearing and vestibular deficits, as the inner ear fluid is common to both systems.
Migraine is also associated with SHL, presumably through vascular mechanisms. Migraine is very common (15% of the population), far more so than SHL, and even occasional SHL due to migraine could be very significant. The author's opinion is that migraine is a much more common cause of SHL than is generally appreciated in the otolaryngologic community. On the other hand, migraine is often just another name for severe headache of unknown cause, which makes this association somewhat meaningless.
Serum antiendothelial cell antibodies are associated with a poorer prognosis (Cadeni et al, 2003). This suggests an autoimmune component. We have encountered a single patient with relapsing polychondritis, who had serial SHL, resulting in complete deafness and loss of vestibular function as well. This is rare.
Pregnancy is not a risk factor for SHL (Yen et al, 2016). This is surprising as pregnancy is a risk factor for another cranial nerve syndrome -- Bell's palsy, and also because there are opportunities for large changes in CSF pressure during delivery. However, perhaps it relates to the fact that pregnancy is often reduces Migraine substantially. The cases of SHL that occur in pregnancy are rare, and generally case reports. (Hou & Wang, 2011; Pawlak-Osinska, Burduk, & Kopczynski, 2009) It is hypothesized that it is due to a hypercoagulable state. (Kanadys & Oleszczuk, 2005; Lavy, 1998) Some cases are attributed to ligamentous laxity. For these situations, hearing loss associated with delivery is the more common situation. (Whitehead, 1999). As SHL is rare, and pregnancy is common, it would also seem reasonable that the two combine purely by chance.
Cerebrospinal fluid (CSF) leak is a particularly interesting causal variant of SHL. CSF leaks may be caused by diagnostic or therapeutic procedures in medicine such as. lumbar puncture, spinal anesthesia or epidural anesthesia (Johkura et al. 2000). Symptoms may occur weeks after the procedure (Lybecker and Anderson, 1995). CSF leak may occur spontaneously and may follow trauma. While CSF leak is generally accompanied by an orthostatic (upright only) headache, this association is not universal and in fact, hearing loss may be more common than headache (Oncel et al, 1992). Orthostatic tinnitus is also possible.(Arai, Takada et al. 2003). The hearing loss of CSF leak likely results from lowering of CSF pressure, which lowers perilymphatic (inner ear) pressure, and results in a picture similar to Meniere's disease. (Walsted et al., 1991). Fortunately, the hearing loss is generally temporary. The treatment is with blood patch.
PLF (perilymphatic fistula) has been gaining ground as a cause of SHL, mainly due to reports from multiple German institutions concerning finding a PLF in about 1/3 of patients with SHL. See the section below on surgical treatment for PLF.
Hypertrophic pachymeningitis of the IAC was recently reported by Muelleman et al (2018) as a very rare cause of SHL.
Bottom line: we favor audiometry, MRI of the posterior fossa with gadolinium, CBC, sed-rate in all persons with SHL not otherwise explained, and additional testing decided based on historical features. In persons who are sexually active, we also favor a test for syphilis (RPR or FTA). Practically, in the US, this is almost never productive.
In essence, SHL is diagnosed by documenting a recent decline in hearing. This generally requires an audiogram.
One can get a quick idea if one has SHL by using a cell-phone hearing app. These have improved immensely in recent years. These apps might also be useful for following ones response to treatment (or no treatment as it may be).
Other studies are performed mainly to look for specific causes. Evaluation usually begins with a careful history looking for potential infectious causes such as otitis media and exposure to known ototoxic medications. Autoimmune hearing loss is suggested by good recovery, response to steroids, and relapse. As hearing in SHL generally improves no matter what you do, including nothing, the observation of relapse is important here.
The "official" recommendation is that clinicians should not routinely do laboratory tests. These sorts of recommendations are sometimes driven by cost control considerations of public health officials, which oddly enough seem to be applied only to diagnostic testing, but usually leave out the immensely expensive treatments. The US health-care system will routinely pay gigantic amounts of money for medications. One wonders why it is OK to pay $30,000/month for a cancer medication, but not OK to spend $100 on a blood test.
Recommendations against routine testing seem to us to be no recommendation at all.
Tests worth considering in SHL
Tests worthwhile in special cases
Tests unlikely to be helpful
Mattox and Simmons (1977) reported a rate of 65% spontaneous recovery to "functional hearing levels." Byl also reported a recovery rate of about 69% (Byl, 1984). Those that recover 50% of hearing in the first 2 weeks following SHL have a better prognosis than those who do not recover at this rate (Ito et al, 2002).
Recurrence of SHL is rare but possible (Furohashi et al, 2002). Harkonen et al (2016) reported cumulative recurrence rate of about 3.5% with an average follow-up. of 8 years. Pecorari et al (2020) reported higher numbers (about 10% at 5 years), but their study was small with only 50 patients having 5 year followup. Of course, recurrence might be interpreted as the wrong diagnosis (e.g. Meniere's is the logical alternative).
Cvorocic et al recently reviewed the prognosis of SHL (2008). Using step-wise discriminant analysis, they reported that a "recovery value" was predicted by the following formula.
R=0.968-.216*Severity-.231*Vertigo+.211*speed of treatment+.113*other ear-.064*audiogram shape
It is better to have a minor hearing loss, no vertigo, and rapid treatment (within 1 week). Less important features are hearing in the other ear and the pattern of the hearing test.
If you have experienced a SHL, you might be able to follow your hearing loss using a cell-phone hearing app. These have improved immensely in recent years.
Occasionally, unfortunate persons experience hearing loss on the opposite ear, a year or more later. Wang et al (2016) reported on 14 of these "successive BSSNHL" patients, with hearing loss in the opposite ear averaging about 11 years after the first. They suggested that the hearing impairment in this select group is more severe and outcome of treatment is worse. Considering this unlikely but devastating possibility, we think it is prudent to take the more aggressive approach in treating SHL (see below).
Bottom line: At this writing (2017) almost everything looks either ineffective or slightly effective. Because hearing tends to recover spontaneously at such a high rate, treatment is not always felt necessary, especially when impairment is minor. There is also an substantial possibility of bias, as it is difficult to "control" procedures that involve injections through the ear drum.
Nevertheless the prospect of being permanently deaf in one ear is daunting and has prompted many trials of therapy. Many persons opt for treatment, hoping to get the small "edge" reported in most studies, but without huge hopes.
We presently suggest that most people start on oral steroids quickly (preferably same day), check and follow their hearing with a cell-phone hearing app, and if the hearing loss is severe, consider the more aggressive steroid regimens outlined below.
This recommendation differs from those of the "Clinical Practice Guideline: Sudden Hearing Loss (Update) Executive Summary. (2019)" in that it is much more aggressive. In the guideline, the suggestion is that hearing testing be obtained within 14 days, and that an MRI as well as IT steroid treatment should be offered "when patients have incomplete recovery from sudden sensorineural hearing loss 2 to 6 weeks after onset of symptoms "
Without any proof, we think that SHL that occurs over seconds is more likely to be vascular, and SHL that occurs over hours is more likely to be viral. We would be inclined to use treatments stratified by this logic, but the data is just not there to say that this approach is any better than any other.
When a treatment of SHL is used, it often consists of burst of steroids such as prednisone. There is an immense variability in otology/neurotology practice in regards to use of management and steroids for SHL (Coelho et al, 2011). 26% preferred oral steroids alone and 22% a combination of oral and intratympanic steroids. Some also used intravenous steroids.
Eisenman and Arts reviewed the topic of steroid treatment (2000). Evidence to date for a good effect is generally mixed. Two meta-analysis studies of steroid treatment (Conlin and Parnes, 2007; Labus et al, 2010) suggested there was no benefit. In a more recent evidence based review, Lawrence and Thevasagayam (2015) suggested that either oral or intratympanic steroids should generally be offered. Even more recently, Gao and Liu (2016), in another meta-analysis, suggested that combined intratympanic and systemic steroids provide better outcomes than systemic steroids alone.
Accordingly most (generally uncontrolled) studies suggest a better hearing prognosis for treated vs. untreated patients (Haberkamp and Tanyeri, 1999; Alexiou et al, 2001; Chen et al, 2003; Slattery et al, 2005; Jeyakumar and Francis, 2006), but a few, a worse prognosis (Minoda et al, 2000).
These studies just never stop coming, and here are a few more:
- Alexiou et al (2001) stated that a better prognosis was associated with very high doses of intravenous prednisolone. This makes some sense as one would think that more would be better, but also have more side effects.
- Huy and Sauvaget, 2005 suggested that hearing outcome is not altered differentially by IV steroids administered in the first day vs. within the first week . This study was probably underpowered as it violates common sense.
- A recent meta-analysis concluded that there was no significant effect (Labus et al, 2010).
- Nakache et al (2015) compared oral steroids to intratympanic injections and suggested that there was no difference (Nakache et al, 2015).
- Jung et al, ( 2016) combined "systemic" and IT steroids suggested a better result with the combined protocol . The systemic steroids included 5 intravenous administrations of dexamethasone. The effect size was about 10 dB. This study was uncontrolled and unblinded, as is usually the case.
- Ashtiani et al (2017) reported no difference was reported between oral steroids, intratympanic steroids, and combination oral+intratympanic. Rather unusually, placebos injections were used. A positive response was determined to be 10 dB. There was no "true placebo" of either injection of salt-water alone, or oral/intravenous placebo.
- Suzuki et al (2018) reported that there was no difference between 2 and 4 intratympanic injections of dexamethasone.
- Sugihara et al (2018) reported that the freqency that steroid injections are administered has no effect on hearing outcome. They were comparing 3 interval timing groups: 1-4, 5-10, and 11-30 days apart. If there is no difference between a 1-4 and an 11-30, as well as no difference between 2-4 injections, it would seem to us that this is not a strong treatment.
Our impression from the literature is that systemic (i.e. oral or IV) as well as intratympanic steroids have a small positive effect on SHL. As of 2017, the best approach for hearing appears to be more steroids - -both systemic and intratympanic. Our guess is that larger and more direct doses of steroids (i.e. intravenous or high-dose prednisone) are slightly better than safer and lower doses of steroids (i.e. medrol dose pack). Lets be clear however -- these are not big effects, and one should be careful to judge whether the risk of steroids is warranted by the small effect. For example, one would generally not want to use high dose systemic steroids in a diabetic, as the risk from steroid effects on diabetes might easily outweigh the small average effect on hearing. This involves an apples/oranges comparison however. The equation might be different in a professional musician with diabetes.
Gianoli reported a good response to intratympanic steroids, in persons who were unable to tolerate oral steroids (Gianoli, 2001). Many others, in uncontrolled studies, have made made a similar reports (Banerjee and Parnes, 2005; Gouveris, and Selivanova, 2005; Slattery and Fisher, 2005; Plaza and. Herraiz 2007; Haynes et al, 2007; Van Wijck and Staecker 2007; Lee and Choi, 2015; Ermutlu et al, 2017). There is an obvious trend for an increasing number of positive reports, albeit nearly all uncontrolled, over time.
One wonders why there are no controlled studies ? Perhaps it is because investigators prefer not to perform a "placebo" injection. This is very understandable. Or perhaps this is because controlled studies show no effect, and studies that show no effect are hard to publish.
Ng et al (2014) performed a metaanalysis of intratympanic steroids and reviewed 187 randomized trials, but found only 5 that met their inclusion criteria. They concluded that in the 5 surviving studies, there was a benefit. Lets look more closely.
Studies are suggestive of good results but most have obvious flaws:
There has been a "controlled" study, and also a recent placebo controlled study. Xenellis, J., N. Papadimitriou, et al. (2006) reported a "rescue" approach where intratympanic steroids were used after 10 days of intravenous steroids. They used patients as their own controls. They reported a statistically better effect in the IT patients, with no change in the controls. We are hopeful but a little dubious. The problem with this study is that it is not a protocol likely to be helpful clinically, as it is not presently common to give 10 days of intravenous steroids in the treatment of SHL. Another concern with this study is that it is implausible in that conventional thought is that treatments provided after 10 days of illness are intrinsically unlikely to work.
A double-blinded study of only 60 patients split into 3 groups (Battaglia et al, 2008) indicated that patients treated with a combination of intratympanic dexamethasone and high dose steroids are more likely to recover hearing than those treated with high dose steroids alone. Placebo IT injections were used in one arm. The steroid regimen involved prescribing 66 tablets of prednsone (10 mg) given in a dose of 6 tablets for 7 days, then 5 capsules for 2 days, then 1 less capsule per day until finished. IT steroids or placebo were administered once/week for a total of 3 weeks. This study suggested a powerful treatment effect where the combination group did far better than groups with either IT dexamethasone alone or high dose steroids + placebo injections. This protocol also would be expected to have numerous steroid side effects due to the prolonged use of high-dose prednisone.
This trial was stopped prematurely due to slow subject accumulation. Because of this it may be underpowered and the conclusions may reflect random statistical events combined with the known tendency of journals to publish "positive" results.A concern that we have with this trial is that the placebo arm involved 4 injections of saline through the ear drum. It seems to us that the injections themselves might have an adverse effect on hearing (compared to oral steroids given without a placebo injection). In other words, this study needs to be repeated, preferably with different subject groups - -combination vs. IT dexamethasone vs. oral alone, as realistically these groups are the one that a clinician might choose.
Similar in concept, but without any controls, In 2015, Lee and Choi, compared (retrospectively) two groups of 229 patients -- one group of 99 that was given combined "systemic steroids" and IT dexamethasone (daily), and another that were given "systemic" steroids alone. The abstract does not define "systemic" -- i.e. intravenous ? oral ? how much ? Furthermore, the protocol of daily IT injections is unusually aggressive. If this didn't work, then they were given IT steroids (called "salvage"). There was roughly a 20% difference in results in favor of the combined treatment (77.8% improved), compared to the systemic steroids along group (60.8%). The obvious problem here that makes interpretation impossible, is the lack of definition of what was done as well as the lack of controls.
Dallan et al (2011) reported that intratympanic steroids had identical results to a mixture of oral steroids, pentoxifylinne, low molecular weight heparin and vitamin E. While Dallan suggested that this result meant that intratympanic steroids were recommended as "first line therapy", in our view, their result suggested that intratympanic steroids was equivalent to a mixture of oral steroids plus placebo (e.g. vitamin E). Ermutlu et al (2017) compared IT steroids to oral steroids across 35 patients (not very many). IT steroids were a little worse in efficacy in this uncontrolled small study.
Kim et al (2015) compared use of both IT and oral steroids to patients with either one alone -- this was an unblinded retrospective study -- thus the level of evidence is low. Combined treatment was accompanied by "significantly higher recovery rates" in patients without hypertension or diabetes. We find this conclusion very dubious, as Kim et al did not say how much better recovery was, and also because their study sounds like a "fishing expedition" (because they excluded hypertension or diabetes when they reported results).
Suzuki et al (2016) reported that it doesn't matter whether steroids are given once/week for a month or 4 times a week. These are both fairly aggressive steroid regimens.
Huang et al (2020) compared two variants of steroid therapy (IT) and IV for 12 days and IT for 12 days. Thus both were, over 24 days (which is a very long duration). They found "There was no evidence of a difference in hearing outcomes between IT dexamethasone and sequential IV plus IT treatments." This trial has no placebo control, and also asks a question that does not really seem especially salient.
Thus overall, the evidence doesn't seem to support IT steroids being more effective than oral steroids, or IV steroids being better than IT steroids-- perhaps neither one makes much difference -- but presumably IT steroids are a little safer, especially in persons with diabetes. Our guess is that neither one is very effective, but we would expect that IT would be a little better due to higher concentrations of the drug. We would also expect that ongoing meta-analysis combining all of these small trials would be needed to draw any firm conclusion. Unfortunately, as reported by Ng et al (2014), the quality of these studies (only 5/187 were OK) is generally not very high.
Here are a few obvious problems with the intratympanic dexamethasone treatment approach:
The data is clearly not strong enough right now to make steroid injections the "standard of care" for all SHL. We would put this instead as an equivocal cost/benefit ratio with low cost (in terms of complications), but unfortunately also low benefit (in terms of efficacy).
Bottom line: We recommend IT steroids in persons who cannot or prefer not to be given oral steroids, and in whom the time frame is short (i.e. 4 days from onset), and in whom the hearing loss is very significant. We do not recommend IT steroids otherwise -- for example, in persons with minor hearing loss. We are not sure ourselves whether oral or IT steroids are superior in terms of hearing results to doing nothing, but we are certain that large amounts of oral steroids can cause significant side effects. Concerning intravenous steroids, they are likely superior to oral steroids, but they are far more difficult to administer. Again, their use would seem most reasonable in persons with very severe hearing loss.
In a person whose hearing improves to a useful level during administration of steroids, and then relapses after steroids are stopped, ongoing immunosuppressant therapy should be considered (such as etanercept). Detection of this pattern requires monitoring of hearing past initial treatment. This doesn't happen very often, and there is an obvious fallacy in that if hearing tends to improve anyway, improvements cannot necessarily be attributed to steroids.
Antivirals seem reasonable, given the frequency that herpes family viruses have been associated with SHL. Nevertheless, studies do not show that they work (Conlin and Parnes, 2007). In a more recent evidence based review, Lawrence and Thevasagayam (2015) suggested that antivirals should not generally be offered routinely. This was also the opinion of the most recent AAO consensus statement (Chandrasekhar et al, 2019). Of course, this leaves a lot to discretion.
In a recent animal study, combination treatment with an antiviral (acylovir) and steroids reduced damage in animals whose ears were inoculated with herpes simplex virus type 1 (HSV-1) (Stokroos, 1999), compared to treatment with either acyclovir or prednisolone alone. Similar results were found in a human study by Zadeh et al (2003). On the other hand, several groups using good methodology and substantial numbers of patients have reported no benefit of Valacyclovir or Acyclovir plus steroids over steroids alone (Tucci et al, 2002; Uri et al, 2003; Westerlaken et al. 2003) and as mentioned above, a meta-analysis showed no effect (Conlin and Parnes, 2007).
Medications like acyclovir or valacyclovir may be unhelpful when the cause is a virus that is not in the herpes family, and one rarely knows at the time of the hearing loss which if any virus is responsible. It is also possible that this sort of treatment is just too late in the course of the disorder, as the average time to treat in the Tucci et al study was 4 days (2002).
SHL is a very disturbing experience, especially when accompanied by tinnitus and vertigo, and there have been many questionable treatments advocated. Basically, people get upset and sometimes make bad decisions. Here, we suggest extreme caution on both part of patients and doctors.
There are several protocols involving increasing blood flow or oxygenation: We think that the oxygen part of these treatments is a placebo. In a recent evidence based review, Lawrence and Thevasagayam (2015) suggested that "The cost, limited availability and lack of strong evidence for hyperbaric oxygen therapy (HBOT) make it impractical at present."
In a bizarre position paper published in 2012, the American Academy of Otolaryngology guidelines for SHL suggested that hyperbaric oxygen might be offered with 3 months of diagnosis of SHL (Stachler et al, 2012). The same group (Chandrasekhar et al, 2019), recently pulled back from this recommendation and instead said "hyperbaric oxygen therapy remains an option but only when combined with steroid therapy". Thi sis a bit more reasonable.
Fattori et al (2001) suggested that hyperbaric oxygen therapy was the treatment of choice. This involved 10, 90-minute sessions of breathing pure oxygen at 2.2 atmospheric pressure in a chamber. Horn et al (2005) also reported some good responses to hyperbaric treatment in an uncontrolled study of 9 patients. Similar results were reported in a larger but still uncontrolled study by Racic, G., S. Maslovara, et al. (2003). Again, Narozny, W., Z. Sicko, et al. (2004) advocate combining steroids with hyperbaric oxygen. This is based on an unblinded and retrospective data. Capuano et al (2015) also reported 84% response in a combined steroid/oxygen group.
Sun et al (2018) compared hyperbaric oxygen to intratympanic dexamethasone for 104 patients with refractory SHL. They reported "There was no significant difference of the total effective rates in the hearing recovery between ITD and HBO group (p = 0.368). However, ITD therapy showed much better improvements of tinnitus than HBO therapy (p = 0.039). After ITD and HBO therapy, there were no significant differences in hearing gains at 2 and 4 KHz between ITD and HBO group (p = 0.468 and 0.934, respectively). Nevertheless, ITD therapy showed significant improvements of hearing gains at 8 KHz (p = 0.049) compared to that of HBO therapy." Overall, from our perspective, it seems that these authors are comparing a placebo (HBO), and a minimally effective treatment (ITD), and found no significant difference. We suspect a larger study might have found a small difference in favor of steroids.
Almosnino et al (2018) again compared hyperbaric oxygen to intratympanic steroids. Similarly to Sun et al, they concluded "The present study demonstrated no significant difference in hearing outcomes between patients receiving salvage therapy with HBO2 and IT steroids compared to patients receiving IT steroids alone."
The author's opinion is that the studies suggesting a positive effect are implausible -- is is difficult to see why hyperbaric oxygen treatment should work and we would not suggest anyone go forward with this until it confirmed with a placebo-controlled and adequately powered study. We would doubt that it would be positive, as we don't see how oxygen, hyperbaric or otherwise, can cause hearing neurons to come back to life. It is well known that just 15 seconds of loss of blood flow causes irreversible hearing loss. Furthermore, it is also well known that inner ear hair cells in humans don't regenerate. A return to life from the dead would be a miracle indeed. It is interesting to note that hyperbaric oxygen is extremely expensive.
Somewhat similarly, but concerning a different method of increasing oxygenation, Mora, R., M. Barbieri, et al. (2003) reported a positive effect of "Intravenous infusion of recombinant tissue plasminogen activator for the treatment of patients with sudden and/or chronic hearing loss." TPA is a powerful anticoagulant and somewhat dangerous, due to bleeding. We are very unenthused about this treatment suggestion deriving from this uncontrolled trial.
Carbogen and MgS04 treatment have also been advocated for SHL (Gordin et al, 2002). These are vasodilators. In a recent evidence based review, Lawrence and Thevasagayam (2015) suggested that "Due to the variability in the vasodilator and vasoactive agents used, there is insufficient evidence to support the routine use of these agents. "
Haberkamp and Tanyeri (1999) noted that while numerous treatments have been studied aiming to improve blood flow, such as carbogen inhalation or stellate ganglion block, all remain controversial or simply lack convincing evidence of efficacy. Very few placebo controlled studies have performed of treatment of SHL and for this reason, there is presently a limited ability to determine what is the optimal treatment of SHL. At this writing we do not feel that there is enough evidence for either treatment for use of treatments intended to increase blood flow.
Recently, Kim et al (2011) compared steroids +carbogen+lipoprostaglandin E1 to several other variants including steroids alone, but not including placebo, They found a small advantage for the combination group. We are uncertain how to interpret this paper. We are dubious about an unblinded study that does not include a placebo arm.
We have had a few patients improve remarkably with migraine treatment (mainly verapamil). No conclusion can be drawn from these occasional good responses. Presumably they had a reversible central hearing disturbance. This cannot be a common situation. We do not think verapamil should be offered to all patients with SHL.
Surgery is NOT the standard of care for sudden hearing loss. Rather, there are a few unusual institutions in Germany, that perform surgery in spite of the general opinion that surgery is not indicated. A general feature of these studies is that there are no non-operated control groups, but also that there is a PLF found about a third of the time.
At the University of Freiburg, exploration of the middle ear and patching is recommended for patients with SHL. According to their article on their experience in 97 patients, surgery can be beneficial if performed within 7 days (Maier et al, 2008). These authors reported that 34% of persons with SHL had a fistula. Their method of judging whether or not a fistula was present was direct observation of the round window while displacing the stapes footplate.
There are several problems with this report. First, it is difficult to follow the logic described by the authors that led them to designated fistula or no-fistula. If the logic is faulty, then their conclusions are also faulty. Second, the natural history of SHL is such that improvement cannot be necessarily attributed to surgical exploration and patching. Without a control group, nothing can be concluded with certainty.
There are similar results reported by several other institutions in Germany.
So in although surgery is not the standard of care for SHL, world wide or in the USA, nevertheless in Germany, exploratory surgery is being done at several hospitals, with some positive results over the last 10 years. It seems likely from these reports that there are some patients with SHL who do indeed have PLF, tentatively on the order of 30%. A meta-analysis would seem in order.
- Similar results are reported in Cologne (Kampfner et al, 2013). In this report, it seems that patients with SHL were explored for fistula. They reported a substantial improvement for operated patients. It is puzzling that these patients were operated, and even more puzzling that they got better. Of course, it would be impossible to blind a study like this.
- Similar results were reported by Loader et al (2017), from Vienna Austria, reporting improvement by better than 10 dB (this isn't much improvement). In this study, there was no non-operated control group, and one wonders what percent would have improved without surgery.
- Similar results were reported by Prenzler et al (2017), at another German Otolaryngology Dept, in Hannover. They reported that 28% of 82 patiens with unilateral profound SSNHL were found to have PLF at the time of operation.
Vitamins, and minerals, and other odd treatments.
Ahn et al. (2006) wrote that Lipo-prostaglandin E1 in combination with steroid therapy is effective for treatment of sudden sensorineural hearing loss in Korean patients with Type 2 diabetes. Even if there is an additional effect of adding prostaglandin, this is a very narrow population and it seems doubtful that this experience generalizes to others. The lack of a control group also is worrisome.
Hatano, M., N. Uramoto, et al, in a paper entitled Vitamin E and vitamin C in the treatment of idiopathic sudden sensorineural hearing loss. (2007) reported a positive effect. Similarly, Joachims, H. Z., J. Segal, et al. (2003) reported a positive effect of vitamin E, when combined with steroids. We think that these positive result are likely to be erroneous and related to the greater ease of publishing positive results over the lack of results. While it is doubtful that vitamin E does any harm, we think it is best to remain cautions given the general lack of efficacy of vitamin E in other contexts. We think it is fine to take vitamin E, but would not criticize care in which it was omitted. A much larger controlled study of these agents would be helpful.
Koo et al (2015) studied injection of Ginkgo biloba extract. Treatment did not improve thresholds. Using Ginkgo to treat hearing loss is a strange idea.
Nageris, B. I., D. Ulanovski, et al. (2004). recommended magnesium treatment for sudden hearing loss based on a small but controlled study in which steroids were combined with either magnesium or a placebo. In our opinion, a much larger trial would be needed to establish efficacy. We also do not see a clear mechanism for this effect.
Xiong et al (2012). This study from China suggests that intravenous administration of a mushroom extract improves hearing in SHL. We are very dubious. Hearing tests are subjective measures, and vulnerable to placebo effect.
Wang et al recently reported that etanercept given acutely in experimental labyrinthitis resulted in much better hearing results (2016). While this animal study may not apply to humans, it suggests that acute treatment with etanercept or a related anti-TNF drug (Remicade, Humira), may improve hearing results for sterile inflammation. See the autoimmune-hearing loss page for more information about these drugs. Of course, a double-blind placebo controlled trial is needed here as well. Practically, the US drug industry has priced these drugs in the nose-bleed range (roughly $2000/dose), and it seems unlikely that this information will do anyone much good until these drugs become generic and cheap.