12 Autoimmune Vestibular Dysfunction, Perilymph Fistulas, Mal de Debarquement Syndrome, and Alcohol-Related Dizziness

Little has changed in these subject areas since the 2008 publication of the first edition of this book, except perhaps some aspects of the management of some of the conditions.1 Despite advances in vestibular assessment over the last decade, the impact on these conditions has yet to be appreciated.

Autoimmune Vestibular Dysfunction

Autoimmune vestibular dysfunction is an immunologically mediated attack on the vestibular system. Autoimmune inner ear disease (AIED)—as the name implies—refers to the involvement of the inner ear. AIED can occur in isolation or in association with systemic autoimmune disease. AIED is uncommon and accounts for less than 1% of all cases of hearing loss and dizziness.²

Apart from the inner ear, the nervous system and other body systems can have immunologic involvement that causes dizziness and disequilibrium. The focus in this section, however, is on the vestibular system, and more specifically the inner ear. Because it is less likely that autoimmune vestibular dysfunction will occur without hearing loss, AIED is reviewed first.

AIED is characterized by rapidly progressive, often fluctuating, bilateral sensorineural hearing loss (SNHL). Tinnitus and aural fullness are often present. Dizziness occurs in 50% of patients with active AIED. Patients are more likely to complain of episodic lightheadedness and mild ataxia than true vertigo.³ Spells of vertigo occur in less than a third of patients. The rapid progression of the hearing loss over weeks to months helps to distinguish it from that attributable to other causes, such as sudden bilateral hearing loss from ototoxicity or noise-induced hearing loss that develops over years. AIED is more common in females between 20 and 50 years old, especially if a systemic autoimmune disease (e.g., rheumatoid arthritis or multiple sclerosis) is present. In 30% of patients with AIED, a systemic autoimmune disorder is present. AIED is rare in children.

There are four theories of how AIED arises: crossreactions, bystander damage, intolerance, and genetic factors. The inner ear, which is not exposed to many antigens, possesses only some active cells, and has no lymphatic drainage, is nevertheless able to generate cellular and humoral immunity. The cochlea has its own innate immunity and the endolymphatic sac can process antigen and produce its own antibody response.⁴ Leukocytes enter the cochlea via the spiral modiolar vein. In AIED, it is believed that the damage due to antibodies or immune cells occurs in certain genetically susceptible patients when they are exposed to unknown environmental pathogens. The damage to the inner ear causes cytokines (e.g., tumor necrosis factor [TNF] and interleukin-1 [IL-1]) to be released in the cochlea. This may also provoke a delayed, additional immune reaction and might explain the attack and remission cycle of disorders like Meniere’s disease. TNF-α expressed by the sac may be one way in which Meniere’s disease is linked to AIED.

Examination of the ear is usually normal. Tuning fork testing may support a diagnosis of SNHL. Patients with dizziness may demonstrate abnormalities, none of which is really diagnostic. In some patients, systemic autoimmune disease may be present. Audiologic testing will reveal progressive bilateral SNHL over time. The hearing may be asymmetric at first and may fluctuate. Otoacoustic emissions (OAE) are likely to support cochlear involvement and electrocochleography (ECoG) may support endolymphatic hydrops. Vestibular testing, including rotatory chair, vestibular evoked myogenic potentials, and platform posturography, may all be used to assess vestibular function. Findings depend on the degree and location of the involvement of the vestibular system.

Imaging does not contribute to the confirmation of the diagnosis in AIED, but MRI should nevertheless be done to rule out bilateral vestibular schwannoma in neurofibromatosis type 2, an unlikely cause for bilateral hearing and vestibular loss.

Serological testing is not essential to make the diagnosis. Anticochlear antibody (also called anti-heat shock protein 70 or anti-HSP 70) also occurs in other disorders, such as Lyme disease. It is also present in up to 5% of the general population. Western blot analysis for this protein may support the diagnosis, but with a sensitivity of ~ 50%, its value is questionable.5 The migration inhibition and lymphocyte transformation tests are not readily available and have not really proved helpful. Other blood tests that may indicate systemic autoimmune disease include erythrocyte sedimentation rate, C-reactive protein, antinuclear antibodies, to rheumatoid factor, antithyroid antibodies, complement C1Q, smooth muscle antibody, antigliadin and antiendomysial antibodies (for celiac disease), HLA testing, and Raji cell assay. Syphilis, diabetes, HIV infection, and Lyme disease may mimic AIED and appropriate tests should be ordered if indicated.

The initial treatment of AIED is the same whether dizziness is present or not. Systemic steroids (prednisone or dexamethasone) are the first line of treatment. A clinical history of rapidly progressive bilateral SNHL over weeks to months and an unrevealing clinical examination, as well as a positive response to systemic steroids, are considered diagnostic for AIED. Prednisone is usually taken daily for 1 month, with repeat audiograms performed after 14 days and 1 month. Regardless of whether or not a patient responds, it is not advisable to continue with a high dose of steroids beyond 1 month. If the steroid is ineffective, it is tapered down and stopped. Patients need to adjust their maintenance dose depending on disease activity; to avoid potential side effects the lowest dose possible should be prescribed.

Since long-term control with systemic steroids is usually not feasible due to side effects, intra tympanic administration, especially of dexamethasone, has been considered as an alternative. This route of administration may reach adequate levels in the perilymph but it is unpredictable and depends on the permeability of the round window membrane. Direct intracochlear administration through a cochleostomy is under investigation. Unfortunately, it needs to be administered on numerous occasions in both ears, necessitating many office visits, which is not practical. Its use in AIED is still off label. Placing ventilation tubes for repeat administration carries a higher risk for permanent tympanic membrane perforation.

After 1 month of systemic steroids, cytotoxic chemotherapeutic drugs are often prescribed, but they have mixed success. Methotrexate is still used either for steroid sparing in responders or as an alternative in nonresponders, despite the fact that it was shown to be ineffective in a large multicenter study.⁶ Cyclophosphamide, a powerful immunosuppressive agent with fast therapeutic activity but with more risks than steroids and methotrexate, should not be used without the help of a clinical immunologist. Plasmapheresis removes humoral and cellular immunopathogens from the circulation, but its potential benefit is relatively short-term. It should be reserved for adjuvant therapy when acute, fulminant AIED associated with systemic immune disease is not responsive to more traditional therapy.

Anti-TNF drugs, such as etanercept and infliximab, show some promise and are currently under investigation.⁷ Ustekinumab is a monoclonal antibody used in the treatment of psoriasis; as an anti-TNF agent, it is directed against IL-12 and IL-23. In general, it seems likely that drugs that work for psoriasis, rheumatoid arthritis, and ulcerative colitis may also be helpful for AIED. Anti-TNF drugs suppress the immune system and may increase the risk of developing lymphoma and other malignancies.

As with other causes of cochlear hearing loss, the goal of replacing the lost and defective cells has stimulated research into stem cell implantation and gene therapy. Fortunately, patients who do not benefit from conventional hearing aids can do well after cochlear implantation. The vestibular dysfunction is usually mild and improves with standard immunotherapy.

It is suggested that 16% of bilateral and 6% of unilateral Meniere’s disease may be caused by immune dysfunction. Bilateral Meniere’s disease may be a clinical variant of AIED. In patients who present with fluctuating hearing loss, aural pressure, tinnitus, and episodic dizziness suggestive of endolymphatic hydrops, a possible autoimmune disorder should be considered.⁸ Another clinical variant of AIED is delayed endolymphatic hydrops. To complicate matters further, the opposite ear can be affected, a phenomenon that has also been described after endolymphatic surgery where an underlying immune-mediated response has been postulated.

The standard treatment regimens for Meniere’s disease are used to treat suspected immune-mediated endolymphatic hydrops (Meniere’s disease); they are discussed in Chapter 7 of this book. The aim is to minimize the dose of systemic steroids required to maintain hearing and to prevent dizziness. Drop attacks are luckily rare in autoimmune disease but often may respond only to surgery. Caution is needed when deciding on a surgical option that destroys vestibular function, as failure of the vestibular function in the nonoperated ear due to the disease process may lead to permanent oscillopsia (Dandy syndrome).

Bilateral vestibular failure (BVF) is a rare condition and in 50% of patients a specific cause cannot be identified. Ototoxicity, Meniere’s disease, and meningitis are the most common identifiable causes. Autoimmune mechanisms are suspected in ~ 5% of cases. Due to low specificity, autoimmune testing is not advised, and in patients with normal hearing, steroids do not improve oscillopsia. Vestibular rehabilitation is currently the only option. Vibrotactile rehabilitative systems may improve general balance but do not restore the vestibulo-ocular reflex (VOR). Vestibular implantation to restore the VOR is currently under investigation.

A perilymph fistula (PLF) is a defect of the oval or round window that produces abnormal communication between the perilymph surrounding the membranous labyrinth and the middle ear space. In his review, Hornibrook highlights the controversy surrounding PLF over the past 50 years.9 Spontaneous fistulas are very rare and controversial and seemingly occur without any obvious antecedent event, but it is now known that the antecedent event is often forgotten and sometimes deliberately not mentioned. Using the most stringent criteria, Meyerhoff and Pollock could not identify any event in at least 2% of proven PLF during surgical exploration in 212 patients.¹⁰ The term “idiopathic fistula” may be a better description for this group.

Acquired fistulas due to an antecedent event also include those iatrogenically created during surgery. The pressure force at the oval and round membrane creating the tear can be implosive or explosive. Implosive fistulas arise from increased pressure in the middle ear due to barotrauma, which can be caused by events like rapid airplane descent, blunt trauma to the ear, whiplash, acoustic trauma, and scuba diving. Implosive forces drive the membranes of the oval or round windows inward, tearing them and permitting escape of perilymph into the middle ear space. Explosive fistulas arise from increased intracranial pressure resulting from activities like weight lifting, vigorous coughing, straining, or the Valsalva maneuver. Increased intracranial pressure is theorized to communicate with the perilymphatic space by way of the internal auditory canal or cochlear aqueduct. Explosive forces drive the membranes of the oval and round windows outward, tearing them and permitting escape of perilymph into the middle ear space. Barotraumatic “implosive” forces are most likely to cause a round window fistula, and head trauma is most likely to cause an oval window fistula. The distinction between implosive and explosive is not particularly important, since symptoms and management are similar. In some cases, a “double membrane break” occurs when, in addition to oval and round window membrane rupture, the intralabyrinthine membrane of the cochlear duct also ruptures.

PLF can result from ear surgery, including stapes surgery, chronic ear surgery, and cochlear implantation (CI). Temporal bone anomalies and inner ear malformations increase the likelihood of fistula formation.¹¹ PLF should be distinguished from a “gusher,” which is a term used for the sudden drainage of profuse clear fluid on making an opening into the inner ear. This fluid is CSF and is due to a defect of variable size between the malformed inner ear and the internal auditory canal (IAC). The onset of the CSF drainage can be delayed after surgery.

Because of indistinct, nonspecific features of the history and physical examination, PLF often represents a diagnostic challenge. Other disorders can present in a similar way and therefore it is important to establish a possible precipitating event in the history. The most common complaints include dizziness, imbalance, chronic positional disequilibrium, and episodic vertigo. Nausea and vomiting may accompany intermittent spells lasting anything from a few seconds to hours. Motion intolerance can occur.

Permanent or fluctuating hearing loss, tinnitus, and aural pressure can occur without vestibular complaints. The evidence that PLF is a cause of sudden hearing loss is poor, unless it is preceded by a distinct traumatic event.

The otoscopic examination can be normal, although tympanic membrane perforation, middle ear effusion, and signs of ossicular dislocation may be present. An office fistula test should be performed by gently applying positive-pressure insufflation followed by negative-pressure insufflation of the ear canal with a pneumatic endoscope. The test is positive when the patient complains of increased dizziness (or induced nystagmus with Frenzel glasses). In the absence of clinical middle ear or mastoid disease, a positive fistula test is referred to as the Hennebert sign. Unfortunately, the test is positive in only half of patients with presumed fistula. The Romberg and Fukuda stepping test may demonstrate unilateral hypofunction on the affected side.

Audiometric testing is nonspecific and may reveal a conductive, sensorineural, or mixed hearing loss. The hearing can be normal. Sensorineural hearing loss may fluctuate.

Nystagmography may show spontaneous nystagmus or positional nystagmus, as well as reduced caloric function in the affected ear, or it may be normal. A fistula test performed with impedance audiometry and recorded with nystagmography is more objective than office testing. Valsalva and Tullio testing used in the diagnosis of superior semicircular canal dehiscence (SSCD) may also be positive. Sheppard and others showed a diagnostic specificity of 56% for PLF in patients who demonstrated postural sway with sinusoidal ear canal pressure stimulation on platform posturography (Video 12.1).¹²

The role of vestibular evoked myogenic potentials (VEMP) in PLF has not been established. Electrocochleography (ECoG) has been reported to aid in the diagnosis; in that regard, a change on the intraoperative ECoG is considered the most unequivocal evidence that a window fistula is present.¹³ Unfortunately, it requires special equipment that is not widely available.

Defects are repaired by connective tissue packing of the involved window and sealed with fibrin glue if available. In 1989, a questionnaire on PLF management was sent to members of the American Otological Society and the American Neurotological Society; 75% of respondents said they would graft a window even if a fistula were not found.14 This is probably still true today.

The House Ear Clinic reported on the exploration of 86 ears over a 12-year period.¹⁵ When a fistula was found, 68% had an improvement in their major symptom; but when a fistula was not found, 29% felt better anyway, suggesting a placebo effect. In patients in whom a fistula was found, one-third had no history of ear surgery or trauma. Based on this, the House Clinic advocated a very cautious approach to the diagnosis of PLF, especially for sudden hearing loss and in children. When the predominant symptom of suspected PLF is hearing loss, recovery of hearing after surgery is rare.

Mal de Debarquement Syndrome

Mal de debarquement (MDD), also known as “sickness of disembarkment,” is a disorder characterized by a persistent feeling of dizziness and disequilibrium that usually follows an ocean cruise. It can also occur after air travel, prolonged train rides, space flight, and even skiing. It is rare, with a prevalence estimated at 0.05%, and it can be very distressing.¹⁶ MDD should be distinguished from other forms of motion sickness, such as land sickness. Land sickness is common and occurs in between 47% and 73% of persons disembarking from seagoing voyages.¹⁷ The persistence of motion, however, is short-lived and resolves spontaneously within 2 days, whereas MDD lasts longer than 1 month, and may persist for more than a year.

In 1796, Erasmus Darwin described some of the symptoms of MDD after travelling by boat and stagecoach.¹⁸ MDD has always been considered a variant of motion sickness. Motion sickness or “mal de mer” is commonly experienced while traveling by boat, but in the majority of patients with MDD, symptoms are not experienced until after disembarking.⁶ On the contrary, it is often reported that travelling afterward reduces the symptoms of MDD.

Patients with MDD complain mainly of a persistent sensation of swaying, rocking, or bobbing immediately or shortly after cessation of the voyage. Spinning vertigo is uncommon. In severe cases, patients may complain of disorientation, impaired cognition, fatigue, ataxia, insomnia, headache, anxiety, and depression. In a case series of 27 individuals, Hain et al reported that 93% of subjects indicated rocking and 81% indicated swaying, with imbalance less common (74%).19 They further indicated that the mean duration of symptoms was 3.5 years, with a range of 1 to 10 years. Symptoms were constant in 85% and intermittent in 15%. MDD mainly affects women: according to a 2014 survey of the Chicago Dizziness and Hearing database, 93 out of 109 patients (85%) were females between 30 and 50 years old.¹⁶

The pathomechanism of MDD is still controversial. The predominant opinion is that it is a variant of motion sickness. On the other hand, motion sickness does not really explain the female and age predominance in MDD; therefore, some experts believe that it is related to migraine. The female preponderance has also raised the question whether genetics (the two X chromosomes) and female hormones can perhaps contribute.

Some experts still believe that MDD is a somatization disorder or a form of anxiety. There are some reports of MDD following use or withdrawal from serotonergic medications. Serotonin may inhibit glutamate, which is an excitatory transmitter in the vestibular nucleus. The data supporting the hypothesis that MDD is caused by reweighting of visual, vestibular, or somatosensory input are contradictory. Nachum et al hypothesized that there is an increased reliance on somatosensory input after motion exposure, and a reduced weighting of vision and vestibular input.²⁰ In contrast to this, Peterka suggested an increased reliance on visual and vestibular information (and thus decreased somatosensory weighting).²¹ This occurs in normal subjects who are exposed to situations where somatosensory feedback is distorted.

Currently the two most attractive proposed mechanisms for the development of MDD are the inappropriate internal predictive model and maladaptation of the VOR to the roll of the head during rotation. An internal predictive model is a method whereby someone reacts to an event before it happens.²² It appears that after a few days on a boat, a person develops an internal model of the periodic motion in his brain. With this model, he predicts and cancels out visual or somatosensory input that is phase-locked to pitch rotation, and enhances responses due to surge that is not. This predictive model of boat motion is selected and applied to avoid falling. When back on land, this model is usually disposed of within hours to days; but in patients with MDD, it takes months and even years to disappear. Patients then benefit from this model only when in motion, for instance when driving in a car. The rest of the time they are symptomatic.

Dai et al recently proposed that MDD was caused by maladaptation of the VOR to roll of the head during rotation.²³ This maladaptation was previously produced in humans in NASA space flight experiments.²⁴ In monkeys, it was shown that only those with long VOR time-constants developed abnormalities, implying that the maladaptation of the VOR depends on the velocity storage mechanism. Velocity storage is the central vestibular mechanism that allows peripheral labyrinthine responses that fatigue with sustained rotational stimulation to be prolonged. It ultimately allows the ability of the VOR to transduce the low-frequency component of head rotation. Maladaptation adds vertical and horizontal components to ocular torsion induced by head roll and body oscillations at a frequency centered on 0.2 Hz.

The diagnosis of MDD is based on the development of symptoms after a sea voyage or prolonged exposure to motion, an improvement when driving, and the exclusion of other vestibular pathology. Motion exposure is on the order of 7 days, with 2 hours being a minimum. The clinical examination, audiometry, and imaging studies are usually unremarkable. If symptoms occur after a flight, perilymphatic fistula should be considered.

There are no vestibular test findings specific to MDD. In a case series by Baloh, quantitative vestibular testing in patients with MDD revealed no consistent abnormality, although a static direction-changing positional nystagmus was noted in 50%.²⁵

Treatment of MDD is still predominantly medical and is mostly ineffective. As stated by Hain, it is merely aimed at making the patient comfortable while awaiting spontaneous remission.¹⁶ Despite potentially being addictive, benzodiazepines, such as low-dose clonazepam, seem to help the majority of patients. Antidizziness and motion-sickness drugs, including meclizine, Dramamine (Prestige), scopolamine, betahistine, baclofen, and verapamil, are of little to no benefit. Anecdotal reports supported the use of some antimigraine drugs, such as gabapentin, amitriptyline, and venlafaxine. SSRI-type antidepressants, dopamine agonists, phenytoin, carbamazepine, and even nonsteroidal anti-inflammatories have been reported to be helpful in selected cases. In 2013, Cha et al, in a pilot study, proposed repetitive transcranial magnetic stimulation (TMS) over the dorsolateral prefrontal cortex to offer short-term symptom improvement.²⁶ Evidence supporting the benefit of physical (balance) therapy is lacking. Although movement by self-motion or driving is often beneficial, it is not advised as treatment and may even prolong symptoms. Surgery plays no role.

Dai et al reported that re-adaptation of the VOR relieves the symptoms in MDD.²³ Using a full-field optokinetic stimulus, given while the head was rolled at the frequency of the subjects’ rocking (usually around 0.2 Hz), reversed the MDD in 23 out of 24 patients. Unfortunately, the treatment effects regressed in six patients. In the other 17 (70%), the MDD was cured or substantially reduced for prolonged periods (mean follow-up = 11.6 months). This approach is currently under further investigation.

Acute alcohol intoxication is associated with gait imbalance, slurring of speech, and, at times, vertigo. Nausea and vomiting often accompany a sensation of impending doom. The stereotypical gait ataxia and dysarthria are suggestive of cerebellar involvement, although vestibular pathways are probably involved as well. The direct effect of alcohol on vestibular nuclei has been noted in animal studies, where synaptic transmission was impaired.27 Vestibular testing in patients with alcohol intoxication has revealed normal vestibuloocular reflex (VOR) gain, albeit with impaired fixation suppression of vestibular nystagmus. This is consistent with cerebellar dysfunction.²⁸ With moderate alcohol ingestion, slowing of saccades and smooth pursuit eye movement is consistently observed.

Gaze-evoked nystagmus is commonly observed with alcohol ingestion. This nystagmus is predominantly observed with horizontal eye movements, with the fast phase of the nystagmus in the same direction as the direction of gaze. This is a reliable sign of intoxication, the magnitude of which is highly correlated with blood alcohol concentration.²⁹

Positional nystagmus, and the associated vertigo, is another effect of alcohol on the vestibular system. This phenomenon has been well studied, particularly with regard to amount, type, and rate of alcohol ingestion.³⁰ In 1911, Barany described the direction-changing characteristics of positional alcohol nystagmus (PAN) in humans with changes in head position.³¹ The specific gravity of alcohol is less than that of endolymph. When alcohol blood levels approach 40 mg/dL, alcohol diffuses into the cupula via its adjacent vascular supply. This makes the cupula lighter than endolymph, transforming the semicircular canals into receptors that are sensitive to gravity. Vertigo and nystagmus then occur in the supine position. In the first phase of alcoholinduced nystagmus (PAN I or resorptive phase), starting ~ 30 minutes after ingestion of alcohol, a geotropic nystagmus is seen, with the fast phase of nystagmus toward the lower ear. This phase may last 3 to 4 hours. The nystagmus is suppressed by visual fixation. This is an important point, as centrally mediated nystagmus is not fixatable, supporting that this phenomenon is peripheral in nature. After PAN I, there is a “silent intermediate period” where there is neither positional nystagmus nor vertigo. This takes place 3 to 5 hours after alcohol ingestion and occurs as alcohol diffuses into the endolymph, with the specific gravities of the cupula and endolymph approaching one another.

The next phase, referred to as the reduction phase or PAN II, occurs 5 to 10 hours after alcohol ingestion, and the direction of nystagmus is opposite that of PAN I, with nystagmus toward the upper ear. This is the case because the specific gravity differential is such that the endolymph becomes “heavier” than the cupula with alcohol moving out of the cupula and into the endolymph. This period is when “hangover vertigo” may be experienced. Positional vertigo may persist until the alcohol completely leaves the endolymph but may not occur until hours after the blood alcohol level has reached zero. PAN II is usually associated with motion sickness and is a major contributor to the hangover. The “morning after” drink thus may have a physiological basis, transiently lessening the intensity of the hangover. The practical implication of PAN is recognized and respected in the aviation industry, where airline pilots are not permitted to fly within 8 to 12 hours of consuming alcohol. The question is whether this “throttle to the bottle” rule should not be extended to 48 hours, in light of PAN.

Chronic alcohol intake leads to cerebellar atrophy and Purkinje cell loss, particularly in the anterior vermis.³² The Purkinje cells regulate and coordinate motor movements by means of inhibition of certain neurons. Wernicke’s encephalopathy, which occurs in alcoholics, is due to a thiamine deficiency. It consists of a triad of acute confusion, ataxia, and ophthalmoplegia and may be fatal if untreated. The ataxia is predominantly related to acute vestibular hypofunction. This syndrome may progress to Korsakoff’s syndrome, which is characterized by a memory deficit in which the person is unable to establish new memories. Patients are noted to confabulate, and this behavior may be used to cover up gaps in memory.³³

Some patients may report a temporary improvement in their dizzy symptoms after consuming small amounts of alcohol.³⁴ In these cases, the possibility of phobic postural vertigo (PPV), chronic subjective dizziness (CSD), or what is nowadays classified as persistent postural perceptual dizziness (PPPD) should be considered.³⁵ In these conditions, alcohol acts as an anxiolytic with which patients self-medicate.

Alcohol is highly toxic to the inner ear when absorbed through the round window, a fact to be taken into account with local application of alcoholcontaining eardrops and solutions.

Stay updated, free articles. Join our Telegram channel

Join