Pathophysiology of tinnitus
Introduction
Tinnitus is a percept that is well characterized in humans and also in some animal models. Here the approach to the pathophysiology of tinnitus will be to start with human tinnitus sufferers and to find differences in cochlear and brain function with those that do not suffer tinnitus. The available methods range from otoacoustic emissions to probe the cochlea, evoked potentials or evoked magnetic fields to reflect synchronous brain activity, and forms of functional imaging such as positron emission tomography (PET) and functional magnetic resonance imaging (fMRI) that are largely reflecting the number of active neurons and their firing rates (Mukamel et al., 2005). A caveat in all these investigations is that tinnitus is most frequent in individuals with hearing loss, and this in itself causes changes in the cochlea (by definition) and in the brain (Don et al., 1998; Oates et al., 2002). Age typically introduces additional changes in the central nervous system (Tremblay et al., 2003) as does gender (Don et al., 1993). Thus, the tinnitus and non-tinnitus groups should preferably be matched for hearing loss, age and gender. Another caveat is that individual differences in evoked potential amplitude and functional images are often so profound that it is impossible to interpret the small changes observed. Therefore, studies where individuals serve as their own controls (unilateral tinnitus, gaze-induced tinnitus that can be controlled) are popular but the results may not be generalized easily. Furthermore, tinnitus as a percept is likely to depend on activity in the auditory cortices, even if the tinnitus generating structure is more peripheral. It is therefore useful to distinguish the tinnitus generation site from the site with the initial pathology, i.e., hair cell loss in the cochlea. We will call the tinnitus generation site, i.e., the site that is most peripheral in the auditory nervous system and that shows an accepted neural correlate of tinnitus, the ignition point (Eggermont, 2006). Here we have temporally neglected the important role of centrifugal cortical activity that, with some time delay, can cause changes in more peripheral structures. We will, as many others have done, equate the strength of the tinnitus sensation with the amount of deviance of the studied correlate from the norm. Correlates could for instance be the amplitude of evoked potentials, the size of the change in activated brain regions, or the incidence of spontaneous otoacoustic emissions.
PET studies in gaze-induced tinnitus suggest various correlates such as regional cerebral blood flow (rCBF) increase in the temporal-parietal auditory association areas (Brodman area: BA 22, 40, 42) but not in primary auditory cortex (PAC) (Giraud et al., 1999) and in addition activation in the auditory lateral pontine tegmentum (Lockwood et al., 2001). Absence of activation in PAC rules out significantly increased spontaneous firing rate (SFR) but does not rule out increased neural synchrony as a correlate of tinnitus. In fMRI studies of gaze-induced tinnitus (Cacace et al., 1996) foci of activity in the tinnitus patient were localized in the superior colliculus and frontal eye fields in cortex. Presently it is not clear whether this reflects activity related to the motor act (gaze) that induces tinnitus or to the percept of tinnitus itself. In patients able to modulate the loudness of their tinnitus by oral facial movements the loci of activity were in the auditory cortex (AC) contralateral to the ear in which the tinnitus was perceived as well as in limbic structures (Lockwood et al., 2001).
Lidocain infusion, which temporally suppresses tinnitus, produced equally likely increases or decreases in rCBF, but the changes were always larger for increases in the right temporal lobe of the auditory association cortex (Reyes et al., 2002), and this unilateral activation pattern suggests that the tinnitus originates centrally rather than peripherally. This is because for unilateral stimulation of a normal hearing subject there are always comparable changes in both hemispheres (Reyes et al., 2002). PET studies in groups of tinnitus patients with either Lidocain or with suppression of tinnitus by masking suggest that the baseline cortical activity in the right hemisphere regardless the lateralization of the tinnitus can be abnormally high in individuals with tinnitus (Mirz et al., 1999).
Using fMRI to determine the effect of tinnitus on the activity in the inferior colliculus (IC), Melcher et al. (2000) found a specific effect of masking noise: in unilateral tinnitus binaural noise produced an abnormally low further activation in the IC contralateral to the tinnitus ear. Since the amount of activation was calculated by comparing the brain oxygen level dependent (BOLD) response in the noise-masking condition to that without masking (but with tinnitus), one can interpret this by assuming that the tinnitus produces increased activation of the IC and that additional noise can only produce very little extra activation in the IC. Alternatives are that the noise masks the tinnitus and thus reduces its loudness and consequently its effect on IC activity. This finding suggests that neurons in the IC in individuals with tinnitus have a high-spontaneous activity. The authors noted that increased neural synchrony without increased SFR could not produce this effect.
Evidence for structural changes in individuals with tinnitus was found exclusively in the right posterior thalamus (including the medial geniculate body, MGB) where gray matter concentration was increased (Muhlau et al., 2006). This could point to increased dendritic arborization, potentially accompanied by increased synaptic density and the potential for increased spontaneous activity. It is interesting that the right auditory cortex, to which the right MGB projects, showed larger increases in activity compared to the left one.
The generation of tinnitus in many individuals is associated with decreased input to the auditory system caused by hearing loss. Various sources of evidence indicate that deprivation of primary auditory input leads to a slow-wave mode functioning of the central nervous system, i.e., analysis of electro-encephalogram (EEG) signals shows an enhanced power in the delta frequency range (<4 Hz). An example of such a condition is slow-wave sleep, during which thalamic (and thus also cortical) centers are partially cut off from input from prethalamic relays. This condition leads to a hyperpolarization of thalamocortical cells that activates sodium and potassium currents, gradually depolarizing the cell. The depolarization in turn triggers a calcium-mediated low-threshold spike burst. The frequency of this hyperpolarization spike-burst cycle is approximately in the delta to theta frequency range. Because of the cortico-thalamic connections, coherent slow-wave oscillations are also seen on a cortical level. The spontaneous cortical neuronal activity recorded as the magneto-encephalogram (MEG) in a group of individuals with tinnitus is characterized by a marked reduction in alpha (8–12 Hz) power together with an enhancement in delta (1.5–4 Hz) as compared to a normal hearing control group. This pattern was especially pronounced for temporal regions. Moreover, correlations with tinnitus-related distress revealed strong associations with this abnormal spontaneous activity pattern, particularly in right temporal and left frontal areas (Weisz et al., 2005a).
Ever since it was suggested that phantom limb pain was associated with changes in the topographic map in sensorimotor cortex (Flor et al., 1995), and that similar map changes in AC could underlie tinnitus (Mühlnickel et al., 1998) the search has been on for an unambiguous demonstration of such map changes. Most of the techniques used are based on the MEG and the maps are constructed on basis of equivalent dipole source locations for a series of frequencies. Albeit that this technique is fraud with pitfalls for those based on the N100 component because of contributions from multiple areas with opposite tonotopic gradients (Lütkenhöner et al., 2003b), it nevertheless suggest that consistent changes do occur. It appears that mapping based on middle latency responses (MLRs) that originate in PAC is more reliable (Lütkenhöner et al., 2003a). Diesch et al. (2004) using the steady-state evoked magnetic field that originates exclusively in PAC found enhanced activity in individuals with tinnitus compared to normal hearing controls. The enhancement correlated with the perceived intensity and intrusiveness of tinnitus. Wienbruch et al. (2006) used the 40-Hz auditory steady-state response (SSR) that also originates in PAC to compare the tonotopic frequency representations between individuals with chronic tinnitus and hearing impairment and normal hearing controls. The SSR frequency gradients were attenuated in both hemispheres in individuals with tinnitus. Dipole power was also elevated in tinnitus, suggesting that more neurons responded synchronously to the 40 Hz sound envelope. The altered frequency representations in tinnitus may reflect a loss of intracortical inhibition in partially deafferented frequency regions of the PAC after deprivation of input caused by cochlear injury. It is of course not clear if the map changes would have been different in a group of individuals with the same hearing loss but without tinnitus. The intricate interaction between hearing loss and tinnitus makes this difficult to study.
Attias et al. (1993) found that in individuals with tinnitus the event-related potentials (ERPs) (N100, P200 and P300), which originate largely from secondary and association cortices, were reduced in amplitude compared to what it was in hearing loss and age-matched controls. This could point to increased spontaneous activity in those areas so that stimulus-evoked activity can recruit fewer non-refractory neurons. Weisz et al. (2005b) compared individuals with tinnitus with controls and used tonal edge-frequency stimuli and tonal stimuli with a frequency one-octave below the edge. They found that the N100 dipole strength for individuals with tinnitus and controls was not different for edge-frequency tones but that the N100 responses were significantly larger for tonal stimuli with one-octave below the edge-frequency in the right hemisphere of the individuals with tinnitus. The source location for the edge-frequency dipole was abnormal, but in contrast to the Mühlnickel et al. (1998) findings, the deviation from the control position was not related to tinnitus distress. This suggests that map changes in itself cannot explain the strength of the tinnitus percept. The data from Weisz et al. (2005b) do suggest increased neural synchrony (larger N100), in contrast to the interpretation of the Attias et al. (1993) data, and enhanced right hemisphere activity as in the PET findings of Reyes et al. (2002) and Mirz et al. (1999). The differential hemispheric activity, with the strongest activity unrelated to the side of tinnitus is not explained by any bottom-up modeling of tinnitus (Eggermont and Roberts, 2004) and does suggest mechanisms potentially related to the differential representation of fast periodic (left hemisphere) vs. slow modulated or continuous (right hemisphere) sounds (Zatorre and Belin, 2001).
Gerken et al. (2001) did not find differences in the auditory brainstem response (ABR), except for the latency of wave VII, but they did observe exceptionally large amplitude of the MLR in their “problem tinnitus” group compared to a hearing loss group without tinnitus and an elderly group (where also large MLRs were found but not as consistent as in the tinnitus group). This could point to either increased neural synchrony, reduced central inhibition a known aspect of aging (Willott et al., 1997), or both.
Tinnitus can be considered as a positive symptom disorder characterized by “neuronal hyperactivity due to loss of afferent inhibition”, and has been linked to bursting neural activity in the medial thalamus. In one study, over 2000 single units (SUs) were recorded in the medial thalamus prior to medial thalamotomy in 104 patients, 6 of which had tinnitus. Approximately 99% of the 2000 SUs were unresponsive to sensory stimulation or motor activation, while 41% of the total showed rhythmic (at ∼4 Hz) or random burst-firing typical for low-threshold calcium spike bursts (Jeanmonod et al., 1996). These authors suggested a linkage between some forms of tinnitus and EEG spindling which is initiated by rhythmic bursting in the thalamus.
Spontaneous otoacoustic emissions are low-level sounds emitted by the healthy normal ear (outer hair cells) that are recordable with sensitive microphones inserted in the ear canal. In ∼6–12% (or at least 4% according to Penner and Jastreboff, 1996) of normal hearing persons spontaneous otoacoustic emissions are considered at least partially responsible for the tinnitus (Norton et al., 1990; Penner, 1990). Plinkert et al. (1990), elaborating on a proposal by Kemp (1981), speculated that the pathological long-term movements of a small local group of not more than 60 affected outer hair cells may account for tonal tinnitus. In most cases, however, spontaneous otoacoustic emissions and tinnitus are independent phenomena (Wilson and Sutton, 1981; Penner and Burns, 1987; Penner, 1992). Although spontaneous emissions could theoretically produce increased “spontaneous firing” in neurons innervating the basilar membrane at the emission site, central nervous system adaptation may preclude their audibility. Occasionally, however, some individuals have been reported to hear intermittent spontaneous otoacoustic emissions as intermittent tinnitus (Burns and Keefe, 1992).
In a patient with an overdose of salicylate, transient evoked otoacoustic emissions (TEOAE) were absent and distortion product otoacoustic emissions (DPOAE) were reduced but still present despite a 50 dB hearing loss, but the input-output functions were linearized suggesting a loss of outer hair cell functioning (Janssen et al., 2000). In individuals with tinnitus with normal hearing, DPOAE amplitudes were also reduced compared to normal hearing controls (Ozimek et al., 2006). Twenty-four hours after involuntary noise trauma in military personnel, TEOAE amplitudes were reduced when the tinnitus appeared to be long lasting and were a better predictor for persisting tinnitus than just the amount of hearing loss (Nottet et al., 2006). In contrast, in tinnitus resulting from head trauma in individuals with normal audiograms, the TEOAE amplitudes were increased, the incidence of spontaneous acoustic emission was doubled, and contralateral tinnitus suppression was reduced, all pointing to a potential deficit in the medial olivocochlear bundle activity (Ceranic et al., 1998). Thus, otoacoustic emissions are useful in the delineation of some mechanisms involved in cochlear functioning that might accompany tinnitus, but they do not generally relate to tinnitus itself.
Head and neck injuries are a profound cause of tinnitus, and in addition head and neck contractions can also modulate existing tinnitus in 80% of patients. More importantly in 60% of patients with no tinnitus at the time of testing such contractions could induce tinnitus even in the profoundly deaf (Levine et al., 2003). Møller et al. (1992) showed that median nerve stimulation could modulate existing tinnitus in 40% of ears, and Rubinstein (1993) showed that 33% of tinnitus patients could modulate their tinnitus with jaw movements. This suggests an important multi-modal aspect in the perception of tinnitus and points to different anatomical structures along the extra-lemniscal (non-classical pathways, see chapter 1) pathways, such as the external nucleus and cortex of the IC, and including the dorsal cochlear nucleus (DCN) as important ignition sites in the generation of tinnitus.
Studies in humans have shown abnormalities such as elevated delta waves in the spontaneous spectrum of the EEG or MEG, relatively enhanced activation in the right hemisphere, increased evoked-potential or evoked-magnetic field amplitude, tonotopic map changes in AC, structural changes in the right-sided thalamus and increased SFRs in the IC. Head trauma induced tinnitus may differ from that induced by noise trauma and the former may have a peripheral component. Multi-modal effects suggest involvement of the extra-lemniscal pathway, pointing to DCN, the external nucleus of the IC (ICx) and secondary auditory cortex (AII) as important potential ignition sites for tinnitus.
Section snippets
Tinnitus inducing agents: etiology
Causes of tinnitus in humans can be catalogued on basis of epidemiological and clinical studies. Henry et al. (2005) list overviews from several studies suggesting that noise trauma is the single most unique cause of tinnitus (18%), followed by head and neck trauma (8%) and ear, nose and throat (ENT) infections and illnesses (8%), whereas drugs only account for 2% of known incidents of tinnitus.
Tinnitus is thus often related to hearing loss resulting from external causes such as noise trauma
Putative neural correlates of tinnitus
The human studies reviewed above suggest changes in SFR (DCN and IC) and in neural synchrony (PAC) as correlates for tinnitus. Increased neural synchrony will be reflected in increased evoked potentials (or local field potentials, LFPs) as well as increased correlation in spiking times between simultaneously recorded neurons. Potentially other abnormalities in spontaneous firing such as increased burst firing cannot be excluded at this point. Another potential correlate of MEG studies suggested
Morphological and anatomical correlates of noise trauma
Exposure of chinchillas with a one-octave band of noise centered around 4 kHz and presented for 105 min at 108 dB sound pressure level (SPL) showed axonal degeneration in the DCN that peaked at 16 days post trauma (Bilak et al., 1996; Morest et al., 1998). However, in the auditory nerve (AN) and ventral cochlear nucleus (VCN) axonal degeneration continued for up to 8 months. This degeneration process was accompanied by an increase in small-diameter axons by up to 90% at 8 months post exposure (
Ion channels
The inner hair cells in the cochlea are equipped with only one type of Ca2+ channels, namely the L-type. These calcium channels regulate the release of glutamate from the inner hair cells. Blocking these L-type channels with nimodipine results in a decrease in spontaneous and stimulus-driven firing rates in AN fibers (Robertson and Paki, 2002). In the IC, salicylate appears to block this L-type Ca2+ channel (Liu et al., 2005). If salicylate would do the same in the inner hair cells, then one
Tonotopic map changes
Local mechanical damage to the cochlea (Robertson and Irvine, 1989; Rajan et al., 1993), ototoxic damage to the cochlea (Harrison et al., 1991), and noise-induced hearing loss all cause tonotopic map changes in PAC (Eggermont and Komiya, 2000; Seki and Eggermont, 2002; Noreña et al., 2003). The map changes are not causally related to the hearing loss (Noreña and Eggermont, 2005) and can occur in the absence of hearing loss as measured by ABR (Noreña et al., 2006) but are always accompanied by
What is the neural correlate of tinnitus?
In the bottom-up studies of tinnitus increased SFR and, to a lesser extent, neural synchrony have attracted most of the attention. It is doubtful whether increased SFRs in subcortical structures (DCN, IC) will lead to increased SFR in cortex. It is far more likely that increased neural synchrony in subcortical structures propagates along the auditory pathway (Kimpo et al., 2003) and ultimately result in increased synchrony and/or SFRs in cortex. Human studies using fMRI that find increased
Abbreviations
- AAF
anterior auditory field
- ABR
auditory brainstem response
- AC
auditory cortex
- ACh
acetylcholine
- AChR
acetylcholine receptor
- AII
secondary auditory cortex
- AMPA
alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid
- AN
auditory nerve
- AVCN
anterior ventral cochlear nucleus
- BA
Brodman area
- BOLD
brain oxygen level dependent
- CAP
compound action potential
- CF
characteristic frequency
- ChAT
cholineacetyl transferase
- CN
cochlear nucleus
- DCN
dorsal cochlear nucleus
- 2-DG
2-deoxyglucose
- DPOAE
distortion product otoacoustic emission
- EEG
Acknowledgements
This study was supported by the Alberta Heritage Foundation for Medical Research, a Canadian Institutes of Health Research — NET grant, and the Campbell McLaurin Chair for Hearing Deficiencies.
References (139)
- et al.
Detection of glutamate decarboxylase isoforms in rat inferior colliculus following acoustic exposure
Neuroscience
(1999) - et al.
Auditory event related potentials in chronic tinnitus patients with noise induced hearing loss
Hear. Res.
(1993) - et al.
Type 1 vanilloid receptor expression by mammalian inner ear ganglion cells
Hear. Res.
(2003) - et al.
Noise-induced cell death in the mouse medial geniculate body and primary auditory cortex
Neurosci. Lett.
(2005) - et al.
Effects of chronic salicylate on GABAergic activity in rat inferior colliculus
Hear. Res.
(2000) - et al.
Differential expression of N-methyl-d-aspartate receptor in the cochlear nucleus of the mouse
Neuroscience
(1996) - et al.
Age-related changes in GABA(A) receptor subunit composition and function in rat auditory system
Neuroscience
(1999) - et al.
Age-related changes in the response properties of cartwheel cells in rat dorsal cochlear nucleus
Hear. Res.
(2006) - et al.
Central auditory aging: GABA changes in the inferior colliculus
Exp. Gerontol.
(1995) - et al.
Effects of acoustic trauma on dorsal cochlear nucleus neuron activity in slices
Hear. Res.
(2002)
Salicylate-induced abnormal activity in the inferior colliculus of rats
Hear. Res.
Salicylate and quinine selectively increase spontaneous firing rates in secondary auditory cortex
Hear. Res.
Moderate noise trauma in juvenile cats results in profound cortical topographic map changes in adulthood
Hear. Res.
The neuroscience of tinnitus
Trends Neurosci.
Auditory evoked responses in control subjects and in patients with problem-tinnitus
Hear. Res.
Reorganization of auditory cortex after neonatal high frequency cochlear hearing loss
Hear. Res.
Direct projections from the dorsal column nuclei and the spinal trigeminal nuclei to the cochlear nuclei in the cat
Brain Res.
Effects of intense tone exposure on choline acetyltransferase activity in the hamster cochlear nucleus
Hear. Res.
Related activity in the dorsal cochlear nucleus of hamsters previously tested for tinnitus following intense tone exposure
Neurosci. Lett.
Intense sound-induced plasticity in the dorsal cochlear nucleus of rats: evidence for cholinergic receptor upregulation
Hear. Res.
Plasticity of spontaneous neural activity in the dorsal cochlear nucleus after intense sound exposure
Hear. Res.
Age-related loss of the GABA synthetic enzyme glutamic acid decarboxylase in rat primary auditory cortex
Neuroscience
Effects of salicylate on serotoninergic activities in rat inferior colliculus and auditory cortex
Hear. Res.
Salicylate blocks L-type calcium channels in rat inferior colliculus neurons
Hear. Res.
Salicylate induced changes in outer hair cell lateral wall stiffness
Hear. Res.
Localization of primary auditory cortex in humans by magnetoencephalography
Neuroimage
Studies of tonotopy based on wave N100 of the auditory evoked field are problematic
Neuroimage
Spontaneous activity in the inferior colliculus of CBA/J mice after manipulations that induce tinnitus
Hear. Res.
Effects of lidocaine on salicylate-induced discharge of neurons in the inferior colliculus of the guinea pig
Hear. Res.
GABAA receptor binding in the aging rat inferior colliculus
Neuroscience
Age-related reduction of [3H]strychnine binding sites in the cochlear nucleus of the Fischer 344 rat
Neuroscience
GAD levels and muscimol binding in rat inferior colliculus following acoustic trauma
Hear. Res.
Positron emission tomography of cortical centers of tinnitus
Hear. Res.
The effects of quinine on cochlear nerve fibre activity in the guinea pig
Hear. Res.
Auditory nerve fibre responses to salicylate revisited
Hear. Res.
Changes in spontaneous neural activity immediately after an acoustic trauma: implications for neural correlates of tinnitus
Hear. Res.
Effects of quinine on neural activity in cat primary auditory cortex
Hear. Res.
Aspirin selectively augmented N-methyl-d-aspartate types of glutamate responses in cultured spiral ganglion neurons of mice
Neurosci. Lett.
Regulation of d-aspartate release and uptake in adult brain stem auditory nuclei after unilateral middle ear ossicle removal and cochlear ablation
Exp. Neurol.
Altered glycinergic synaptic activities in guinea pig brain stem auditory nuclei after unilateral cochlear ablation
Hear. Res.
Chemical synaptic transmission in the cochlea
Prog. Neurobiol.
The inhibitory effect of intravenous lidocaine infusion on tinnitus after translabyrinthine removal of vestibular schwannoma: a double-blind, placebo-controlled, crossover study
Otol. Neurotol.
N-methyl-d-aspartate antagonists limit aminoglycoside antibiotic-induced hearing loss
Nat. Med.
Elevated fusiform cell activity in the dorsal cochlear nucleus of chinchillas with psychophysical evidence of tinnitus
J. Neurosci.
Intermittent tinnitus resulting from unstable otoacoustic emissions
Age-related changes in the inhibitory response properties of dorsal cochlear nucleus output neurons: role of inhibitory inputs
J. Neurosci.
Alterations in average spectrum of cochleoneural activity by long-term salicylate treatment in the guinea pig: a plausible index of tinnitus
J. Neurophysiol.
Presence of tinnitus indicated by variable spontaneous otoacoustic emissions
Audiol. Neurootol.
Enhancement of steady-state auditory evoked magnetic fields in tinnitus
Eur. J. Neurosci.
Cited by (105)
Neurophysiological correlates of residual inhibition in tinnitus: Hints for trait-like EEG power spectra
2021, Clinical NeurophysiologyA contribution to the debate on tinnitus definition
2021, Progress in Brain ResearchTinnitus and tinnitus disorder: Theoretical and operational definitions (an international multidisciplinary proposal)
2021, Progress in Brain ResearchChoice of test stimulus matters for pitch matching performance: Comparison between pure tone and narrow band noise
2019, Hearing ResearchCitation Excerpt :Typically, a tinnitus sensation does not present such amplitude modulations; this has to be considered when investigating the performance of patients, suffering from tinnitus, in matching their percept with an NBN. A series of studies (De Ridder et al., 2011; Eggermont, 2007; Møller, 2007) demonstrated that tonal tinnitus is mainly caused by a malfunctioning of neurons in the auditory cortex. However, not a single neuron but rather a population of neighbouring neurons with similar characteristic tuning frequencies is involved in this process.
Changes in auditory thalamus neural firing patterns after acoustic trauma in rats
2019, Hearing Research