Abstract
Otoacoustic emission (OAE) testing enables us to identify the cochlear component of a hearing disorder and to monitor objectively minute changes in cochlear status undetectable by other audiological methods. Contralateral sound-induced suppression is mediated by medial superior olivary complex efferents which induce hyperpolarization counteracting the amplifying effects of outer hair cell (OHC) activity. The aim of this study was to assess functions of cochlea and its efferents in migraine using OAE testing and contralateral suppression of transiently evoked OAEs (TEOAE). Fifty-three migraineurs (106 ears) and 41 healthy subjects (82 ears) were included and pure tone audiometry (PTA), speech discrimination scores (SDS), distortion product OAE (DPOAE), TEOAE and contralateral suppression of TEOAEs were tested. PTA and SDS of migraineurs and controls were not different (P > 0.05). DPOAEs were tested between 1 and 6 kHz and a significant difference was detected only at 5 kHz frequency, where DPOAE amplitudes in migraine with aura (MA) were lower than in controls (P < 0.03). The mean amplitudes of TEOAEs were statistically insignificant between controls and migraine groups. Contralateral sound stimulus induced significant decrease in amplitudes of TEOAE (P = 0.005) in controls. In patients with migraine without aura and MA, mean amplitudes of TEOAEs were not suppressed by contralateral sound stimulus (P > 0.05). As PTA, SDS and DPOAE tests demonstrate normal functioning of inner ear between 1 and 4 kHz, absence of suppression of the TEOAEs by contralateral sound stimulation indicates the presence of dysfunction either in the medial olivocochlear complex in the brainstem or at the synaptic transmission between olivocochlear efferents and OHCs in the cochlea. Disruption in the contralateral suppression may be one of the mechanisms predisposing to the phonophobia symptom associated with migraine headache.
Introduction
Migraine is a chronic, multifactorial neurovascular disorder in genetically susceptible individuals, and alterations in the internal or external environment could initiate migraine attacks (1). Its pathophysiology is complex, and neuronal hyperexcitability may account for the wide range of symptoms of migraine along with aura and headache. Migraine is one of the disabling diseases, and although about 18% of women and 6% of men in the population suffer from the disease, less than half of them are aware that their symptoms are migraine related. In addition to phonophobia, which is the most common auditory symptom, migraine disease can cause a variety of neurotological symptoms such as vertigo, dizziness, hearing loss, tinnitus and aural pain (2, 3). The vast majority of patients experience these symptoms without headache.
Otoacoustic emissions (OAEs) were first described by Kemp in 1978. Since then, OAE testing has been an important clinical and diagnostic tool. OAE testing evaluates the function of cochlea and detects minute changes in outer hair cells (OHCs) undetectable by other audiological methods. The OAEs are acoustical signals that occur spontaneously as narrow-band tonal signals or after stimulation of the ear, and refer to energy being carried backward from the cochlea toward the middle ear. They are produced by active micromechanisms of the OHCs of the organ of Corti. The OAEs are low-level audio-frequency sounds that are produced by the cochlea as part of the normal hearing process. The OAEs can be measured from the external ear canal simply, rapidly and non-invasively.
Transiently evoked OAEs (TEOAEs) can be recorded in almost any subject. A click stimulus is used to elicit a broad spectrum of responses that are most robust in the mid-frequency region between 1 kHz and 4 kHz. The distortion production OAEs (DPOAEs) are the result of an intermodulation distortion produced by the ear in response to two simultaneous, pure tone stimuli that are nearby in frequency and formally referred to as the primary tones. Such a response is described as being distorted because it originates from the cochlea as a tonal signal that is not present in the eliciting two-stimulus complex. The lower frequency tone is referred to as f1 primary, and its level as L1, whereas the higher frequency primary is called f2, and its associated level is L2. The primaries are related in frequency in that the frequency separation of f2 from f1 (termed the f2/f1 ratio) is typically around f1 × 1.2. The most frequently measured acoustic intermodulation-distortion product is at the 2f1–f2 frequency (4). Different frequency regions of the cochlea can be tested by changing primary frequencies.
Olivocochlear activity provides improvement in threshold detection and intensity discrimination of tones in a noisy environment and enhances the responses of auditory nerve fibres to brief tones in the presence of noise (5, 6). Suppression of OAE is mediated by the efferents that originate from the superior olivary complex. Axons of the lateral and medial olivocochlear bundles extend dorsomedially from the superior olivary complex, pass through the reticular formation, combine into a bundle close to the floor of the fourth ventricle, leave the brainstem as a ventral component of the inferior vestibular nerve and join to the cochlear nerve as Oort's vestibulocochlear anastomosis at the base of the modiolus. Axons of the lateral olivocochlear bundles synapse with the afferent (type I) neurons that originate from the inner hair cells in the organ of Corti. Axons of the medial olivocochlear bundles enter the organ of Corti and terminate at the base of the cell bodies of OHCs. The majority of medial efferents project or cross back to the ear of stimulus origin. It is believed that medial efferents induce hyperpolarization that counteracts the contractile (amplifying) effects of OHC activity. Medial efferent stimulation reduces the gain that would result from OHC activity (7). This inhibitory effect is probably mediated by cholinergic transmission (ACh), where calcitonin gene-related peptide (CGRP) or GABA is usually co-released. Since the contralateral sound-induced suppressive effect is mediated by the medial superior olivary complex (MSOC) neurons, contralateral suppression of TEOAEs can be used as a direct index to MSOC efferent activation. The functioning of the MSOC can be tested non-invasively through contralateral suppression of OAEs. Contralateral suppression of OAEs refers to a reduction in amplitude of OAEs at stimulation of the contralateral ear. This effect is attributed to alteration of cochlear micromechanics by MSOC, which can be activated by contralateral acoustic stimulation (8).
OHC cell integrity by OAEs has not yet been investigated in migraine. In this study, we aimed to assess functions of the cochlea and cochlear efferents in the brainstem in migraine using OAE testing and contralateral suppression of TEOAE.
Methods
Patients
Fifty-three migraineurs (43 female, 10 male) were included in the study, and 106 ears were tested. Of 53 migraineurs, 37 had migraine without aura (MoA) and 16 had migraine with aura (MA). The mean ages of the migraineurs with and without aura were 37.2 ± 9 years (range 23–49) and 36.4 ± 12 years (range 15–61), respectively. The diagnosis of migraine was made according to the criteria of the International Headache Society (9). The diagnosis of migraine was compatible with the International Classification of Headache Disorders-II code 1.1 and 1.2.1 and all the MA patients had only visual aura (not accompanied by any other transient neurological deficits). Migraine subjects were not using any prophylactic medication and had not experienced any attack for 3 days prior to that on which they collaborated in the experiment. For exclusion of other diseases, complete blood counting, blood chemistry, thyroid function tests and magnetic resonance imaging (MRI) of the brain were performed. All patients had an intact ear drum on otoscopic examination. The patients had no associated ear problem such as hearing loss, tinnitus or vestibular symptoms. Migraineurs had phonophobia associated with headache, but phonophobia did not exist during the interictal period when the tests were performed.
Forty-one healthy controls without any primary or secondary headache (30 female, 11 male) were included, and 82 ears were tested. The mean age in the control group was 35.2 ± 11 years (range 25–53). None of the controls had chronic otological disease, ototoxicity, ear surgery, central nervous system disease or any systemic, metabolic or autoimmune disease that could impact on ear functions. Informed consent was obtained from all subjects.
Audiometry
Pure tone and speech audiometry (AC 40 Clinical Audiometer; Interacoustics, Assens, Denmark) was performed in all patients and controls. The results were recorded at the frequencies 250, 500, 1000, 2000, 4000 and 6000 Hz. Speech discrimination scores (SDS) and speech reception thresholds (SRT) were obtained. Acoustic impedance tests were performed to evaluate middle ear pressure and presence of ipsi- and contralateral acoustic reflexes.
Otoacoustic emission testing
TEOAEs and DPOAEs were recorded consecutively and analysed utilizing an ILO-88 cochlear emission analyser (Otodynamics, London, UK) using insert ear phones in a sound-attenuated room.
On DPOAE testing, the intensity levels of the primary tones held constant and DPOAE data were recorded for different frequency regions 1, 2, 3, 4, 5 and 6 kHz and plotted as a function of f2. The frequency ratio of the two primary tones (f2/f1) was fixed at 1.22. Stimulus levels were kept at 65 dB for f1 and 55 dB for f2 frequencies. DPOAE measurement at 2f1 − f2 was considered significantly different from the background noise if it exceeded it by at least 3 dB.
The TEOAEs were obtained from 1 to 4 kHz with stimuli consisting of clicks of 80 μs duration. The stimulus level in the outer ear was set at 83 ± 3 dB sound pressure level. The click rate was 50/s, and poststimulus analysis was in the range of 2–20 ms. A total of 260 sweeps was averaged above the noise rejection level of 47 dB. Stimuli were presented in the non-linear mode, in which every fourth click stimulus was inverted and three times greater in amplitude than the three preceding clicks. A TEOAE was defined as a response if its amplitude was ≥ 3 dB above the level of the noise floor. Reproducibility percentages ≥ 70% and stimulus stability ≥ 80% were taken into account as acceptable for analysis.
For contralateral suppression of TEOAEs, insert ear phones (ear tone 3 A 10 Ω 410–3010) were used in the contralateral ear, from where a 40-dB sensation level white noise was given to obtain the optimal measurement parameters (10). Five successive measurements were performed in the frequency range 1–4 kHz, and mean values of the amplitudes obtained with these successive measurements were used in the comparisons. The contralateral suppression test was performed in all migraineurs, but only in 22 ears of 11 controls (eight female, three male).
The Statistical Package for Social Sciences 11.0 program was used for the statistical analyses (SPSS Inc., Chicago, IL, USA). Pure tone audiometry and DPOAE results in MA, MoA and controls were compared using one-way
Results
There was no difference between patients and controls in age and gender. Pure tone averages and speech discrimination scores (P > 0.05) were comparable between non-headache controls and migraineurs and indicated a normally functioning ear (Table 1). The DPOAEs which evaluated the cochlea in a frequency-specific manner did not reveal any abnormality in controls and migraineurs except for the 5 kHz frequency, where DPOAE amplitudes in MA were significantly lower than in controls (P < 0.03) (Table 2).
Pure tone audiometry results of patients with migraine and controls
MoA, Migraine without aura; MA, migraine with aura.
DPOAE testing results in patients with migraine and controls
Statistically significant difference (P = 0.03) between MA and controls at 5 kHz.
DPOAE, Distortion product otoacoustic emission; MoA, migraine without aura; MA, migraine with aura.
The TEOAEs could be obtained in 92% of the controls, whereas the emissions were present in 67.5% of migraineurs (P < 0.01). On
Transiently evoked otoacoustic emissions (TEOAEs) with and without contralateral suppression in patients with migraine and controls. TEOAE amplitudes obtained without contralateral sound stimulation are not statistically different between groups. In healthy controls, the amplitudes of TEOAEs obtained with contralateral stimulation were significantly lower than those of TEOAEs obtained without contralateral stimulation (P = 0.005), demonstrating the presence of contralateral sound-induced suppression in TEOAEs. In MoA and MA patients, the amplitudes of TEOAEs obtained without contralateral suppression were not significantly different from those of TEOAEs obtained with contralateral suppression (P > 0.05). Results indicate that contralateral sound-induced suppression in TEOAEs is impaired in patients with migraine. Results are given as mean ±
Discussion
Vestibulocochlear disturbances can occur as an aura, accompanying the headache or during headache-free periods in migraine (11, 12). Activation of trigeminal sensory innervation of the cochlear blood vessels altering blood flow and vascular permeability, and vasospasm of the branches of the internal auditory artery and even migrainous infarction have been proposed as possible mechanisms for cochleovestibular deficits associated with migraine headache. Subjects included in our study were asymptomatic regarding cochlear or vestibular symptoms such as hearing loss, tinnitus or vertigo. Although patients have phonophobia symptoms associated with the headache phase, they were investigated during the interictal period when they were not experiencing phonophobia. MRI revealed no abnormality in the brainstem, and, as confirmed by pure tone and speech audiometry, no otological involvement at clinical level was detected.
Otoacoustic emissions are invariably associated with functioning OHCs and objectively monitor minute changes in cochlear integrity that are undetectable by other audiological methods. The OHCs are responsible for cochlear sound amplification and capable of generating transient and distortion product otoacustic emissions (TEOAEs and DPOAEs) in response to sound stimulus. DPOAE precisely detects cochlear dysfunction in a frequency-specific manner and is superior in measuring higher frequency ranges (13). DPOAEs tested in a range from 1 to 6 kHz did not reveal any abnormality in MoA patients. However, DPOAEs in MA patients showed lower amplitudes only in the 5-kHz frequency region. As regards the sensitivity of DPOAE for high frequencies, detection of impairment at 5 kHz may suggest an abnormal OHC function in the basal turn of the cochlea in MA. Overall, DPOAE testing indicated a normal functioning cochlea at the frequency range of 1–4 kHz in migraine patients. It has been reported that thresholds for pure tone hearing at high frequencies between 6 and 8 kHz were decreased in migraine patients, despite pure tone hearing not being affected at the 0.5–4-kHz range (14). Pure tone audiometry did not reveal any impairment in our study, although it was not tested for the frequencies beyond 6 kHz. Therefore, decreased DPOAE amplitudes at the 5-kHz region support the contention that there may be cochlear involvement at high-frequency regions of the cochlea in migraine. On the other hand, DPOAE levels found in human studies do not seem to allow for an accurate prediction of the amount of hearing loss (15, 16). Accordingly, a correlation should not be expected between pure tone audiometry and OAE testing results at all times.
TEOAEs originate from the activation of whole cochlea with a click stimulus and have the advantage of evaluating middle-frequency ranges 1–4 kHz. However, limited cochlear lesions in the high frequencies do not significantly alter the amplitudes of TEOAEs. Therefore, both TEOAEs and DPOAEs should be used as they are complementary to each other, and both tests were employed in the present study for cochlear evaluation in migraineurs. The TEOAEs were obtained in 67% and 92% of migraine and control subjects, respectively. This finding might also imply impairment in the activation of whole cochlear OHC cells generating OAEs in migraine. This possibility can be disregarded, since DPOAE were specifically investigated and demonstrated normal functioning OHC cells within the cochlea in the range 1–4 kHz. The amplitudes of TEOAEs in migraineurs seemed lower than those of the control group, although the difference was not statistically significant. Combining the results of DPOAEs and TEAOEs, we can suggest that the cochlea functions normally up to the 5-kHz region in migraine.
Suppression of OAE may also give an indication of the functional integrity of the central mechanisms that control cochlear activity. The medial olivocochlear system as a part of this regulatory complex has particular importance. The effect of the medial olivocochlear system on the cochlea can be assessed by contralateral acoustic stimulation that reduces the electromotility of the OHCs and induces suppression of TEOAEs, particularly in relation to amplitude (Fig. 2). In migraine patients, contralateral sound stimulus did not induce significant suppression of TEOAEs. TEOAE amplitudes in the control group were significantly reduced upon exposure to contralateral sound stimulation. For suppression of TEOAEs, if the criterion of an amplitude decrease of > 1 dB were taken into account, a suppressive effect was obvious in controls where the mean decrease was 1.6 dB in TEOAE amplitude, whereas contralateral sound stimulus resulted in mean 0.7-dB and 0.8-dB decreases in MA and MoA, respectively. The lack of contralateral suppressive effect of TEOAEs during the interictal phase in migraineurs raises several possibilities. First, despite there being no statistically significant difference between the baseline values of TEOAEs of migraineurs and controls, subtle changes in OAE amplitudes may account for the disrupted suppressive effect. However, previous studies showed that no significant correlation was found between the suppressive effect mediated by the olivocochlear system and TEOAE amplitudes (17). In addition, contralateral suppression was performed in subjects who had positive TEOAE (> 3 dB amplitude). Therefore, it is unlikely that lack of contralateral suppression in migraineurs can be attributed to TEOAE amplitudes.
A simplified illustration depicting the olivocochlear pathways involved in contralateral suppression of otoacoustic emissions (OAEs). Transiently evoked OAE (TEOAE) tests outer hair cell function ipsilaterally. Sound stimulation of contralateral ear activates lateral (not shown here) and medial superior olivary complex (MSOC) via anterior ventral cochlear nucleus (aVCN). An indirect projection via medial nucleus of the trapezoid body-medial nucleus of trapezoid body is inhibitory (interrupted lines) on MSOC neurons. Descending efferents from higher brain structures to modulate excitability of MSOC neurons are not shown. MSOC predominantly send efferents to ipsilateral outer hair cells (OHCs), and this synaptic transmission is probably cholinergic, resulting in hyperpolarization of OHCs and suppression of OAEs (interrupted lines). Absence of medial olivocochlear efferent-mediated suppression of TEOAEs in migraineurs implies a dysfunction either (1) centrally in MSOC neurons due to altered balance between excitation and inhibition, or (2) peripherally at cholinergic synaptic transmission between olivocochlear efferent terminals and OHCs. Consequently, our results suggest that decreased olivocochlear efferent suppression interictally may generate susceptibility to noise sensitivity in migraineurs.
The presence of abnormality of OAE suppression indicates impairment in the medial cochlear efferent innervation (18, 19). Contralateral suppression, a reduction in OAE levels following acoustic stimulation of the contralateral ear, is specific to cochlear efferent activity that it is also known as the ‘olivocochlear reflex’ (Fig. 2). The medial cochlear efferent neural pathway that originates in the MSOC, synapse onto the OHCs within the cochlea (20). Direct electrical and auditory stimulation of the efferent pathways reduces OHC function and, in turn, TEOAE amplitudes (21). As changes in OHC function are reflected in TEOAEs, they provide non-invasive means by which the medial efferent neural system in humans can be investigated. The inhibitory effect of the efferent neural system on OHC function (i.e. TEOAE amplitude) as a result of contralateral acoustic stimulation has been shown to disappear or become reduced when there is physical or anatomical disruption of the efferent neural pathways proximal to the cochlea (22).
The MSOC receives excitatory synaptic inputs from anterior ventral cochlear nuclei and inhibitory inputs from the ipsilateral medial nucleus of trapezoid body (23). Inhibitory transmission is predominantly mediated by glycine receptors. Moreover, the excitability of MSOC neurons is also modulated by higher brain centres such as inferior colliculi, cortical and subcortical structures (24, 25), and the total effect of inhibitory and excitatory afferent projections determines their activity. However, the functioning of the rest of the efferent system above MSOC remains obscure, as contralateral sound stimulus activates the olivocochlear efferents. The loss of contralateral sound-induced suppression of OAE in migraineurs may imply an altered balance between excitation and inhibition resulting in reduced activation of the medial olivary complex in the brainstem. Data acquired from various studies are consistent with the notion that general interictal cortical dysfunction towards hyperexcitability exists in migraineurs (26–28).
Alternatively, the dysfunction may arise from synaptic transmission between medial olivary complex efferent fibres and OHC cells, a synapse that has many similarities to the neuromuscular junction. Acetylcholine is the principal efferent transmitter released onto OHCs. CGRP and GABA also occur in olivocochlear efferent terminals and may be co-released with acetylcholine (29–31). Consistent with the major role of acetylcholine within this synapse, targeted deletion of nicotinic acetylcholine subunit abolishes the suppressive effect of medial olivary complex activation (32–35). It has been recently shown that impaired synaptic transmission in myasthenia may cause abnormalities in OAE amplitudes and their suppression mechanisms, which can be partially restored after administration of acethycholine esterase inhibitors (33). This finding in a neuromuscular junction abnormality emphasizes the significance of OAE testing in detecting disruption of cholinergic synaptic transmission in the cochlea. The possibility of subclinical involvement of medial olivary complex–OHC synaptic transmission is not unlikely in migraineurs (Fig. 2), since studies have revealed subclinical neuromuscular transmission impairment in MA patients (36, 37).
Disruption of the mechanisms that control suppression of OAE leads to abnormalities in discrimination, understanding and lateralization of sounds in a noisy environment. Therefore, deficient contralateral suppressive mechanisms in migraine may be associated with abnormal sensitivity to noise as the quality of sounds perceived by the patient may change. Hypersensitivity to auditory stimuli, inducing a feeling of discomfort and avoidance behaviour, is called phonophobia and is one of the characteristic symptoms associated with migraine attacks (38, 39). MoA and MA do not differ as to sound sensitivity, and phonophobia does not correlate with duration, frequency or severity of attacks (38). Being a typical feature of headache, phonophobia has also been reported to occur in approximately 75% of migraineurs in a headache-free period. Migraineurs have also noted that many attacks may be provoked following exposure to a noisy environment (2). Although phonophobia usually coexists with hyperacusis, the latter is used to denote painful sensitivity to sounds and may be accompanied by tinnitus and trauma. Phonophobia and hyperacusis are complex symptoms in which derangements of sensory and cognitive processing mechanisms are implicated. Disturbances of cochlear efferents as detected by impaired contralateral suppression of TEOAEs can lead to auditory symptoms such as hyperacusis, tinnitus and noise intolerance (39). However, the impact of suppression of OAE on the symptom of phonophobia has not been investigated to date. Although our patients were asymptomatic during evaluation, lack of suppression in TEOAEs in migraineurs disclosed a susceptibility to noise intolerance. Understanding of this issue needs further investigation in which OAE suppression in migraineurs will be assessed during both ictal and interictal periods.
Subclinical dysfunction of cochlear efferents in migraineurs is an intriguing finding. Although interictal features of migraine remain mainly unidentified, growing evidence has begun to clarify the interictal characteristics of migraineurs and implies general subclinical nervous system involvement including habituation deficit in cortex and brainstem, cortical dys-excitability, autonomic system dysfunction, impaired neuromuscular transmission and altered neurovascular coupling (40–46). All of those interictal nervous system abnormalities, widely distributed from the cerebral cortex to the neuromuscular junction, would probably be characteristics underlying susceptibility to migraine headache. Although the interictal characteristics of the brainstem have not yet been extensively studied, we believe that dysfunctional central mechanisms modulating the cochlear activity demonstrate another aspect of migrainous brain in handling sensory signals. Consistent with the dorsal pons activation observed during a migraine attack (47), reduced habituation of the nociceptive blink reflex in migraineurs interictally (48) suggests that dysfunction of pons is not restricted to the headache phase and may reflect a common subclinical abnormality underlying migraine susceptibility. In that sense, the presented study also provides evidence for the dysfunctional brainstem hypothesis of migraine headache.
In conclusion, DPOAE testing has disclosed subclinical OHC involvement in high-frequency regions corresponding to basal turn of the cochlea in MA patients. Absence of suppression of the TEOAEs by contralateral sound stimulation in migraine patients may indicate the presence of a dysfunction of medial olivocochlear efferents in the brainstem and/or in synaptic transmission between olivocochlear axons and OHCs in the cochlea. Disruption of contralateral suppression may be one of the mechanisms predisposing to noise sensitivity in migraineurs.