In this article, the present state of the art with respect to audiovestibular testing for Meniere’s disease (MD) is reviewed. There is no gold standard for MD diagnosis, and the classic dictum is that even the “best” tests yield positive results in only two-thirds of patients with MD. Still, we advocate the use and further investigation of advanced audiovestibular testing in patients with MD in an attempt to answer the questions that confront any clinician who cares for patients with audiovestibular symptoms.
One hundred and fifty years after Prosper Ménière described the symptom complex of recurring episodic vertigo, hearing loss, and tinnitus, the diagnosis of Meniere’s disease (MD) remains challenging for the clinician. In part, we are limited by the confines of inner ear diagnostics. Postmortem histopathologic examinations of human temporal bones have shown that many patients with classic MD have distortion and dilation of the endolymphatic spaces of the membranous labyrinth, a finding known as endolymphatic hydrops. These alterations are proposed to be the pathologic basis of MD; however, the correlation between these histologic changes and the clinical manifestations of MD is not absolute. For example, in a double-blind temporal bone study, Rauch and colleagues found histologic evidence of endolymphatic hydrops in 13 of 13 cases of clinical MD, but review of medical records of 6 of 19 temporal bones with endolymphatic hydrops did not reveal symptoms or signs of MD. Thus, many inner ears may have hydropic changes without manifesting the clinical syndrome. Some have suggested that endolymphatic hydrops may be an epiphenomenon of the true, as yet undiscovered, pathophysiologic mechanism of MD.
Without definitive diagnostic criteria for MD, clinicians must rely on historical information and other clinical features to make the diagnosis. At present, by the American Academy of Otolaryngology–Head & Neck Surgery (AAO–HNS) criteria, the diagnosis of MD depends on symptoms of recurrent, spontaneous vertigo, hearing loss, aural fullness, and tinnitus, as well as objective documentation of hearing loss, with other potential causes excluded by proper otologic investigation. The presence and severity of symptoms may fluctuate frequently over time, and physical findings are generally lacking. Thus, there continues to be a strong interest in identifying additional means of testing for MD.
In this article, the present state of the art with respect to audiovestibular testing for MD is reviewed. Because there is no gold standard for MD diagnosis, it is impossible to measure the performance characteristics of the various test methods. The classic dictum is that even the “best” tests yield positive results in only two-thirds of patients with classic MD. Still, we advocate the use and further investigation of advanced audiovestibular testing in patients with MD in an attempt to answer several questions confronting any clinician who cares for patients with audiovestibular symptoms:
- 1.
Does the patient have MD? We review how audiometry can be used to determine if a patient fulfills the AAO–HNS criteria for MD. Audiovestibular testing is also particularly useful for patients who present with only portions of the classic symptom complex. In such cases, objective testing might differentiate a patient likely to have MD with normal hearing at baseline from one with migraine-associated vertigo. Tests may also detect incipient MD, allowing us to initiate nondestructive treatments at earlier stages.
- 2.
Which ear is causing the symptoms? Audiovestibular testing can be used to simplify difficult treatment decisions. If we are able to confirm the diagnosis and lateralize the pathology to one ear, we can proceed with potentially destructive therapy with more confidence.
- 3.
Is there bilateral disease now, or might there be in the future? It is estimated that 30% of patients with MD will develop bilateral disease. It would be helpful to be able to exclude or confirm contralateral disease when considering the treatment approach for the first ear.
- 4.
Is our treatment effective? Once therapy is initiated, it would be desirable to monitor the patient for progression or remission. Expansion of the knowledge of the available and emerging investigative methods may facilitate critical evaluation of treatment regimens and interventions targeted at halting disease progression.
Audiometric testing
Hearing loss in patients with MD is predominantly sensorineural, fluctuating, and progressive. The hearing loss tends to involve the low frequencies early in the course of the disease. Higher frequencies may also be affected, and some have observed a peak, or inverted-V, audiogram in patients with MD. With time, the hearing loss tends to flatten and become less variable. Nevertheless, there is no audiometric configuration that can be considered characteristic of MD, and the configuration does not depend on disease duration. Fluctuation of pure tone thresholds, word recognition, or both is commonly noted. In patients with long-standing MD (>10 years), the average pure tone threshold in the affected ear stabilizes at about 50 dB, and the mean word recognition score reaches a minimum of 50%. Profound hearing loss occurs in only 1% to 2% of patients. The sensorineural hearing loss is cochlear in etiology, with associated distortion, loudness recruitment, and a reduction of word recognition scores in proportion to the pure tone average.
It is interesting to note that a mild low-frequency conductive hearing loss is not uncommonly observed early in the course of the disease. Muchnik and colleagues observed a low-frequency air-bone gap with no middle ear pathology and normal acoustic reflexes in 33% (13 of 40) of patients with classic MD. Although some have dismissed this finding as artifactual, it is reasonable to propose that this is an “inner ear conductive hearing loss,” as a result of increased inner ear fluid volume and pressure (related to endolymphatic hydrops) dampening the stapes footplate mobility. In retrospect, these studies may have included some patients with vestibular symptoms and low-frequency conductive hearing loss secondary to undetected superior semicircular canal dehiscence. The clinician should keep this entity in mind when confronted with such a presentation.
AAO–HNS Guidelines
As defined by the 1995 AAO–HNS Committee on Hearing and Equilibrium, diagnostic guidelines for MD require defined historical criteria (vertigo, tinnitus, aural fullness) and audiometrical documentation of sensorineural hearing loss in the suspect ear on at least one occasion. The hearing loss may take any of the following forms: (1) average thresholds at 250, 500, and 1000 Hz at least 15 dB higher than the average of 1, 2, and 3 kHz; (2) in unilateral cases, the 4-frequency pure tone average of thresholds at 0.5, 1.0, 2.0, and 3.0 kHz is at least 20 dB poorer in the affected ear than in the contralateral ear; (3) in bilateral cases, the 4-frequency pure tone average is greater than 25 dB in the studied ear; or (4) the clinician judges that the patient’s hearing loss meets reasonable audiometric criteria for hearing loss characteristic of MD. A change of 10 dB in the 4-tone average or of 15% in word recognition score is considered clinically significant fluctuation. To avoid the subjectivity and ambiguity inherent in a staging system based on hearing, vestibular, and other symptoms, the Committee also proposed a staging system based solely on hearing as measured by audiometry. All patients will not progress through all stages in sequence, but this system may aid in the analysis of treatment results. The stages are based on the 4-frequency pure tone average, in dB, of the worst audiogram during the 6-month interval before treatment, as follows: (1) 25 or less, (2) 26 to 40, (3) 41 to 70, (4) greater than 70.
Loudness Recruitment
To support a diagnosis of MD, some clinician-investigators have exploited the psychoacoustic phenomenon of loudness recruitment, defined as the abnormal growth of perceived loudness with increasing stimulus intensity. In a series of 200 patients with clinical features of MD, all were found to have recruitment, regardless of the severity of their hearing loss. The most straightforward test of loudness recruitment is the alternate binaural loudness balance test (ABLB). In this procedure, auditory stimuli are presented to both ears, and the stimulus intensity in the suspect ear is adjusted until its perceived loudness is equal to that in the other ear. These results are then plotted as a function of healthy ear stimulus intensity (ie, with the suspect-ear threshold on the x-axis and the other ear threshold on the y-axis). Recruitment is evidenced by a slope function greater than 1. This test is time-consuming and can be challenging for some patients. We do not routinely use the ABLB at our institution, but we often look for indirect evidence of recruitment by history and audiometry (eg, complaints of sound sensitivity in the affected ear, decreased most-comfortable loudness level, narrowed dynamic range, or elicitation of an acoustic reflex at a level that would typically be subthreshold in a normal ear). Although commonly associated with endolymphatic hydrops, recruitment is present in numerous nonhydropic etiologies of cochlear hearing loss, and is therefore not specific for MD.
Vestibular testing
Vestibular test results are not included in the 1995 AAO–HNS Guidelines for the Diagnosis and Evaluation of Therapy in MD, and, strictly speaking, are not necessary for the diagnosis of MD. Although the audiogram remains the most widely used clinical test for MD, it is the vertigo attacks that are often the most disabling symptom of the disease. There is no reason to insist that audiometric changes will directly correlate with vestibular symptoms. Thus, some physicians have argued that vestibular tests can be a useful adjunct in identifying the diseased ear(s) in patients with MD whose symptoms and audiogram are nonlateralizing, and for monitoring the development of early disease in the contralateral ear. Use of vestibular testing in the evaluation of patients with MD remains controversial, however, as these test results often fluctuate during the course of the disease, cannot reliably identify the affected ear, and the degree of damage detected correlates poorly with patient-perceived disability. Currently, the most widespread application of vestibular testing in MD has been to exclude pathology in the contralateral labyrinth in patients who are candidates for ablative therapy.
Caloric Testing
Caloric testing represents a nonphysiologic, low-frequency (∼0.003 Hz) stimulation of the horizontal semicircular canal. It is the only test in the standard vestibular test battery that provides lateralizing information. The definition of a clinically significant caloric asymmetry varies from laboratory to laboratory, but most institutions consider a 20% to 28% or greater canal paresis by Jongkee’s formula to be abnormal. This common vestibular test abnormality is reported in up to 50% to 66% of MD patients. This leaves a significant proportion of patients with MD who will have no evidence of pathology on caloric testing. Furthermore, there does not appear to be a clear relationship between the development of canal paresis and duration of symptoms in the ear with MD. The magnitude of caloric paresis ranges from 25% to 50%, and complete loss of caloric response is rare. Based on temporal bone studies, the mechanism for caloric paresis in MD may be related to ampullary distortion with disruption of the attachment of the cupula.
Use of caloric data to identify the affected ear in MD is highly controversial. Dimitri and colleagues argued that if a 5% interaural caloric difference was used in a forced choice paradigm, this test could correctly identify the diseased ear in 95% of cases; however, most institutions continue to use Jongkee’s formula as a benchmark for pathologic asymmetry. Using these more widely accepted criteria, canal paresis has been documented in the contralateral ear in up to 19% of patients with unilateral MD who underwent serial vestibular testing, which may represent an irritative lesion.
Caloric data are often considered essential before administering any form of ablative therapy in unilateral MD to confirm that the contralateral labyrinth is functional. Complete loss of vestibular function as manifested by absent caloric response to ice water irrigation need not be pursued as a therapeutic end point in intratympanic gentamicin therapy, as it is not required for vertigo control in MD, and is associated with an unacceptably high rate of sensorineural hearing loss in the treated ear.
Spontaneous Nystagmus
The second most common abnormality seen during videonystagmography in patients with MD is spontaneous nystagmus. It is often seen during an acute attack or within days following an attack. Three larger series found that spontaneous nystagmus was present in 20% to 67% of patients with MD, but there is a lack of consensus regarding the criteria for pathologic spontaneous nystagmus. The direction of the observed spontaneous nystagmus varies; it can consistently beat toward the involved ear (irritative), away from it (paralytic), or change from an irritative to a paralytic pattern over time, and thus cannot be used to lateralize the disease. Some authors have attempted to explain the irritative pattern of nystagmus with the membrane rupture theory, arguing that ruptures of inner ear membranes allow the potassium-rich endolymph to transiently leak into the perilymph that bathes the VIIIth nerve and basal surfaces of the hair cells. The resultant potassium intoxication leads to an initial partial depolarization of the vestibular nerve with an increase in the resting discharge rate (irritative lesion). This may then be followed by complete depolarization and inhibition from a blockade of transmitter release (paralytic lesion).
Spontaneous nystagmus can be a useful clinical finding when following patients with MD who undergo intratympanic gentamicin therapy. Minor found that development of spontaneous nystagmus beating away from the treated ear was 1 of 3 clinical signs that specified completion of intratympanic gentamicin therapy, and was associated with a high vertigo control rate. This does not require formal videonystagmography and can be easily done at the bedside with Frenzel lenses.
Rotational Testing
Rotational testing is a natural stimulus for the peripheral vestibular system and relies on angular acceleration. It assesses more physiologically relevant frequencies of the vestibulo-ocular reflex (VOR) compared with caloric testing. It is nonlateralizing. The 2 most common methods to rotate a patient are (1) passive rotation of the head and body in the horizontal plane with rotary chair, or (2) autorotation. This latter method uses active head rotation and can be used in either the horizontal or vertical planes.
Rotary chair test findings in MD are extremely nonspecific and various abnormalities have been described in a small number of studies. Palomar-Asenjo and colleagues found that 23% of patients with MD had elevated phase leads in at least 2 consecutive frequencies on sinusoidal harmonic acceleration. Decreased, normal, and elevated gains have been described in patients with MD. Autorotation may permit assessment of higher VOR frequencies compared with rotary chair, but its test-retest reliability is unclear.
Posturography
Computerized dynamic posturography (CDP) tests postural control in 6 different conditions designed to emphasize or minimize vestibular, visual, and proprioceptive input. It provides useful information about a patient’s functional balance status, but does not directly assess deficits in any of these 3 sensory systems. There is some interest in use of posturography in MD to stage the degree of functional impairment.
Vestibular testing
Vestibular test results are not included in the 1995 AAO–HNS Guidelines for the Diagnosis and Evaluation of Therapy in MD, and, strictly speaking, are not necessary for the diagnosis of MD. Although the audiogram remains the most widely used clinical test for MD, it is the vertigo attacks that are often the most disabling symptom of the disease. There is no reason to insist that audiometric changes will directly correlate with vestibular symptoms. Thus, some physicians have argued that vestibular tests can be a useful adjunct in identifying the diseased ear(s) in patients with MD whose symptoms and audiogram are nonlateralizing, and for monitoring the development of early disease in the contralateral ear. Use of vestibular testing in the evaluation of patients with MD remains controversial, however, as these test results often fluctuate during the course of the disease, cannot reliably identify the affected ear, and the degree of damage detected correlates poorly with patient-perceived disability. Currently, the most widespread application of vestibular testing in MD has been to exclude pathology in the contralateral labyrinth in patients who are candidates for ablative therapy.
Caloric Testing
Caloric testing represents a nonphysiologic, low-frequency (∼0.003 Hz) stimulation of the horizontal semicircular canal. It is the only test in the standard vestibular test battery that provides lateralizing information. The definition of a clinically significant caloric asymmetry varies from laboratory to laboratory, but most institutions consider a 20% to 28% or greater canal paresis by Jongkee’s formula to be abnormal. This common vestibular test abnormality is reported in up to 50% to 66% of MD patients. This leaves a significant proportion of patients with MD who will have no evidence of pathology on caloric testing. Furthermore, there does not appear to be a clear relationship between the development of canal paresis and duration of symptoms in the ear with MD. The magnitude of caloric paresis ranges from 25% to 50%, and complete loss of caloric response is rare. Based on temporal bone studies, the mechanism for caloric paresis in MD may be related to ampullary distortion with disruption of the attachment of the cupula.
Use of caloric data to identify the affected ear in MD is highly controversial. Dimitri and colleagues argued that if a 5% interaural caloric difference was used in a forced choice paradigm, this test could correctly identify the diseased ear in 95% of cases; however, most institutions continue to use Jongkee’s formula as a benchmark for pathologic asymmetry. Using these more widely accepted criteria, canal paresis has been documented in the contralateral ear in up to 19% of patients with unilateral MD who underwent serial vestibular testing, which may represent an irritative lesion.
Caloric data are often considered essential before administering any form of ablative therapy in unilateral MD to confirm that the contralateral labyrinth is functional. Complete loss of vestibular function as manifested by absent caloric response to ice water irrigation need not be pursued as a therapeutic end point in intratympanic gentamicin therapy, as it is not required for vertigo control in MD, and is associated with an unacceptably high rate of sensorineural hearing loss in the treated ear.
Spontaneous Nystagmus
The second most common abnormality seen during videonystagmography in patients with MD is spontaneous nystagmus. It is often seen during an acute attack or within days following an attack. Three larger series found that spontaneous nystagmus was present in 20% to 67% of patients with MD, but there is a lack of consensus regarding the criteria for pathologic spontaneous nystagmus. The direction of the observed spontaneous nystagmus varies; it can consistently beat toward the involved ear (irritative), away from it (paralytic), or change from an irritative to a paralytic pattern over time, and thus cannot be used to lateralize the disease. Some authors have attempted to explain the irritative pattern of nystagmus with the membrane rupture theory, arguing that ruptures of inner ear membranes allow the potassium-rich endolymph to transiently leak into the perilymph that bathes the VIIIth nerve and basal surfaces of the hair cells. The resultant potassium intoxication leads to an initial partial depolarization of the vestibular nerve with an increase in the resting discharge rate (irritative lesion). This may then be followed by complete depolarization and inhibition from a blockade of transmitter release (paralytic lesion).
Spontaneous nystagmus can be a useful clinical finding when following patients with MD who undergo intratympanic gentamicin therapy. Minor found that development of spontaneous nystagmus beating away from the treated ear was 1 of 3 clinical signs that specified completion of intratympanic gentamicin therapy, and was associated with a high vertigo control rate. This does not require formal videonystagmography and can be easily done at the bedside with Frenzel lenses.
Rotational Testing
Rotational testing is a natural stimulus for the peripheral vestibular system and relies on angular acceleration. It assesses more physiologically relevant frequencies of the vestibulo-ocular reflex (VOR) compared with caloric testing. It is nonlateralizing. The 2 most common methods to rotate a patient are (1) passive rotation of the head and body in the horizontal plane with rotary chair, or (2) autorotation. This latter method uses active head rotation and can be used in either the horizontal or vertical planes.
Rotary chair test findings in MD are extremely nonspecific and various abnormalities have been described in a small number of studies. Palomar-Asenjo and colleagues found that 23% of patients with MD had elevated phase leads in at least 2 consecutive frequencies on sinusoidal harmonic acceleration. Decreased, normal, and elevated gains have been described in patients with MD. Autorotation may permit assessment of higher VOR frequencies compared with rotary chair, but its test-retest reliability is unclear.
Posturography
Computerized dynamic posturography (CDP) tests postural control in 6 different conditions designed to emphasize or minimize vestibular, visual, and proprioceptive input. It provides useful information about a patient’s functional balance status, but does not directly assess deficits in any of these 3 sensory systems. There is some interest in use of posturography in MD to stage the degree of functional impairment.
Vestibular Evoked Myogenic Potentials
Vestibular evoked myogenic potential (VEMP) testing is a more recent addition to the MD diagnostic armamentarium. The VEMP is obtained by measuring the relaxation of the sternocleidomastoid muscle (SCM) in response to an ipsilateral auditory stimulus. Brief high-intensity monaural clicks or tone-bursts produce a large, short-latency inhibitory potential (VEMP) in the tonically contracted ipsilateral SCM. Although the exact neural pathway is still not fully clear, the VEMP is considered to be a vestibulocollic reflex with the afferent limb arising from sound-responsive sensory cells in the saccule. The afferent signal is conducted centrally via the inferior vestibular nerve, and efferent signal is conducted via the vestibulo-spinal track to produce inhibitory postsynaptic potentials in cervical motor neurons. Normal responses are composed of biphasic (positive-negative) waves. By convention, they are labeled “p” (for positive) and “n” (for negative), followed by their mean latency value in milliseconds. The first wave complex, present in most healthy participants, is labeled p13-n23.
VEMP testing is a useful adjunct to other studies because it can reveal saccule dysfunction. After the cochlea, the saccule is the second most common site of hydropic changes in temporal bones of patients with MD. It is reasonable to expect that the altered mechanics of a distended saccule might lead to an altered VEMP in MD. This hypothesis has been verified by several studies. Although Cheng and colleagues detected click-evoked VEMP (C-VEMPs) in 98% of normal ears, De Waele and colleagues reported that C-VEMPs were absent in 54% of patients with MD. In another retrospective review, Murofushi and colleagues found that the C-VEMP response was significantly reduced in amplitude or absent in 51% of patients with MD.
Rauch and others have suggested that the diagnostic utility of VEMP may be improved using threshold measures for tone burst stimuli (TB-VEMP). Healthy adults have frequency-dependent TB-VEMP thresholds, with the best response occurring at 500 Hz. In practical terms, a patient has normal frequency tuning if the threshold for a 500-Hz TB-VEMP is less than the thresholds for 250-Hz and 1000-Hz stimuli. In their prospective cohort study of 14 healthy adults and 34 adults with unilateral MD by AAO–HNS criteria, Rauch and colleagues observed that the affected ears of patients with MD had significantly increased TB-VEMP thresholds compared with unaffected ears and healthy subjects. Furthermore, affected ears demonstrated alterations in frequency-tuning (eg, threshold at 500 Hz >at 1000 Hz), and the VEMP threshold did not correlate with ipsilateral audiometric thresholds. Interestingly, the “Meniere’s-like” response was also observed in the unaffected ears of some patients with MD. In a follow-up study, Lin and colleagues found that 27% of patients with unilateral MD showed elevated thresholds and altered tuning characteristics in the asymptomatic ear. This is similar to the 30% rate of second ear involvement reported in the literature, and to the 35% rate of occult saccular hydrops in the asymptomatic ear discovered in their concurrent review of postmortem temporal bones from patients with unilateral MD. They propose that VEMP may be used not only as an adjunct to predict the side of disease, but also as a potential clinical detector of “presymptomatic” hydrops, and thus may be used to predict the development of bilateral disease. Prospective clinical trials are under way in this regard.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
