Tuning Out the Noise: Limbic-Auditory Interactions in Tinnitus
- 1 Laboratory of Integrative Neuroscience and Cognition, Georgetown University Medical Center, Washington, DC 20057-1460, USA
- 2 Department of Neurology, Klinikum Rechts der Isar, Technische Universität München, D-81675 München, Germany
- Available online 23 June 2010
Tinnitus, the most common auditory disorder, affects about 40 million people in the United States alone, and its incidence is rising due to an aging population and increasing noise exposure. Although several approaches for the alleviation of tinnitus exist, there is as of yet no cure. The present article proposes a testable model for tinnitus that is grounded in recent findings from human imaging and focuses on brain areas in cortex, thalamus, and ventral striatum. Limbic and auditory brain areas are thought to interact at the thalamic level. While a tinnitus signal originates from lesion-induced plasticity of the auditory pathways, it can be tuned out by feedback connections from limbic regions, which block the tinnitus signal from reaching auditory cortex. If the limbic regions are compromised, this “noise-cancellation” mechanism breaks down, and chronic tinnitus results. Hopefully, this model will ultimately enable the development of effective treatment.
Since the landmark review by Eggermont and Roberts (2004) on the neuroscience of tinnitus, the level of interest in this often debilitating disorder has mounted steadily. This can be attributed to the rising average age of populations in Western countries and increased hearing loss in young people, partly due to the popularity of music players with in-ear loudspeakers. Tinnitus is also one of the most frequently reported problems among veterans returning from two recent armed conflicts, often co-occurring with traumatic brain injury (TBI) and posttraumatic stress disorder (PTSD). Although the number of published articles on tinnitus is growing and more funding agencies are now supporting tinnitus research, our basic understanding of the disorder is stagnating (Adjamian et al., 2009, Langguth et al., 2007 and Shulman et al., 2009).
It has been assumed for some time that most cases of tinnitus are caused by peripheral noise-induced hearing loss followed by changes in the central auditory pathways (Jastreboff, 1990). Animal models have corroborated this explanation (Irvine et al., 2001, Rauschecker, 1999b and Robertson and Irvine, 1989) but have not provided a conclusive answer as to the location and nature of these central changes (Eggermont and Roberts, 2004). Using a whole-brain approach, human neurophysiological and functional imaging studies have visualized various regions of hyperactivity in the auditory pathways of tinnitus patients (Arnold et al., 1996, Hoke et al., 1989, Lanting et al., 2009 and Melcher et al., 2009) as well as cortical regions beyond classical auditory cortex, including prefrontal and temporo-parietal areas (Giraud et al., 1999, Mirz et al., 1999, Mirz et al., 2000, Schlee et al., 2009 and Weisz et al., 2007). Imaging studies have also demonstrated activation of nonauditory, limbic brain structures, such as hippocampus and amygdala, in tinnitus patients (Eichhammer et al., 2007, Lockwood et al., 1998, Mirz et al., 2000 and Shulman et al., 2009).
This limbic activation has been interpreted as a reflection of the emotional reaction of tinnitus patients to the tinnitus sound (Jastreboff, 2000). As the present article will argue, however, limbic and paralimbic structures may play a more extended role than previously proposed. In our model, efferents from structures in the subcallosal area, which includes the nucleus accumbens (NAc) of the ventral striatum and the ventral medial prefrontal cortex (vmPFC), are involved in the cancellation of the tinnitus signal at the thalamic level. Although the tinnitus signal may initially be generated in parts of the auditory system, it is the failure of the limbic regions to block this signal that leads to the tinnitus percept becoming chronic.
Lesion-Induced Reorganization of the Central Auditory System
Tinnitus, i.e., hearing a disturbing tone or noise in the absence of a physical sound source, is a phantom sensation (Jastreboff, 1990) comparable to phantom pain felt in an amputated limb (Ramachandran and Hirstein, 1998). In both cases, the firing of central neurons in the brain continues to convey perceptual experiences, even though the corresponding sensory receptor cells have been destroyed (Birbaumer et al., 1997 and Rauschecker, 1999a). As such, chronic tinnitus is thought to originate from plastic reorganization of auditory cortex following peripheral deafferentation. According to this “remapping” hypothesis, the reorganization process usually begins with a loss of hair cells in the inner ear, a “sensorineural” hearing loss (SNHL). This cochlear lesion can result from acoustic trauma, i.e., loud-noise exposure within a certain frequency range, or age-related hair-cell degeneration (usually corresponding to high frequencies). Although the lesion causes elevated thresholds in the corresponding frequency range, neighboring frequencies become amplified because their central representations expand into the vacated frequency range. Indeed, preliminary findings from human PET and MEG studies indicate an expansion of the frequency representation in the auditory cortex that corresponds to the perceived tinnitus frequencies (Lockwood et al., 1998, Mühlnickel et al., 1998 and Wienbruch et al., 2006). These observations, however, continue to await confirmation by high-resolution fMRI studies (Leaver et al., 2006).
Animal models of tinnitus also support the claim for a cortical origin of the tinnitus percept. Restricted cochlear lesions in cats and monkeys lead to a frequency-specific reorganization of auditory cortex and thalamus but not of more peripheral stations (Rajan and Irvine, 1998, Rajan et al., 1993 and Schwaber et al., 1993). As detailed microelectrode mapping in these and other studies has shown, frequency regions adjacent to the lesioned area invade the vacated space and become “overrepresented” compared to other frequency regions. Additionally, the “lesion-edge frequencies” lose intracortical inhibitory input from the deafferented region. Thus, cortical neurons with input from frequency ranges next to the cut-off frequency display permanently elevated spontaneous activity (“hyperactivity”) as well as transiently enhanced burst-firing and increased synchronous activity (Noreña and Eggermont, 2003 and Weisz et al., 2006). Auditory brainstem hyperactivity has also been reported in human tinnitus patients (Melcher et al., 2000) as well as animal models of tinnitus (Bauer, 2003 and Brozoski et al., 2002), but it seems to be the cortical remapping that ultimately forms the basis for a chronic tinnitus percept (Melcher et al., 2009).
Lesion-induced plasticity in the adult brain has also been documented in the somatosensory and visual systems (Calford et al., 2005 and Florence and Kaas, 1995) (but see Smirnakis et al., 2005). The cellular and synaptic mechanisms responsible for this process include the unmasking of hidden inputs (Buonomano and Merzenich, 1998) and the sprouting of new connections (Darian-Smith and Gilbert, 1994). Furthermore, changes at the cortical level have been found in combination with perceptual consequences of phantom sensations (Ramachandran et al., 1992) and a “filling-in” of the deprived region by neighboring representations (Gilbert et al., 2001). A similar line of work on the cortical contributions to focal dystonia (a somato-motor disorder affecting specific body parts, such as the hand in musicians or writers) has argued that this disorder shares the same signature (Breakefield et al., 2008, Elbert et al., 1998, Flor and Diers, 2009 and Hirata et al., 2004). Taken together, these findings suggest that lesion-induced reorganization has similar consequences across modalities and that the same process in the auditory cortex may be an underlying cause of tinnitus.
Recently, high-resolution structural MRI studies with voxel-based morphometry (VBM) have indicated that tinnitus-related reorganization of auditory cortex in humans may be accompanied by structural changes at the thalamic level (Mühlau et al., 2006). The VBM technique can identify differences in volume and tissue density of brain regions by direct comparison between two groups. Using VBM on tinnitus patients, Mühlau et al. (2006) observed a significant increase of gray-matter density in the posterior thalamus, including the medial geniculate nucleus (MGN). Using individual morphological segmentation, which is better able to detect changes within small and anatomically variable areas, Schneider et al. (2009) also found volume loss in the medial portion of Heschl's gyrus (HG) in tinnitus patients. Additional evidence from the somatosensory system suggests that lesion-induced functional reorganization may be amplified at the thalamic level, and thus structural changes could manifest more readily there: effects of cortical reorganization of the hand representation in macaque monkeys are enhanced in the ventroposterior nucleus of thalamus via descending projections (Ergenzinger et al., 1998).