Positron emission tomography of cortical centers of tinnitus

Abstract

Tinnitus is associated with a wide variety of disorders in the auditory system. Whether generated peripherally or centrally, tinnitus is believed to be associated with activity in specific cortical regions. The present study tested the hypothesis that these cortical centers subserve the generation, perception and processing of the tinnitus stimulus and that these processes are suppressed by lidocaine and masking. Positron emission tomography was used to map the tinnitus-specific central activity. By subtracting positron emission tomography images of regional cerebral blood flow distribution obtained during suppression of the tinnitus from positron emission tomography images obtained during the habitual tinnitus sensation, we were able to identify brain areas concerned with the cerebral representation of tinnitus. Increased neuronal activity caused by tinnitus occurred predominantly in the right hemisphere with significant foci in the middle frontal and middle temporal gyri, in addition to lateral and mesial posterior sites. The results are consistent with the hypothesis that the sensation of tinnitus is associated with activity in cortical regions functionally linked to subserve attention, emotion and memory. For the first time, the functional anatomy of conditions with and without the habitual tinnitus sensation was obtained and compared in the same subjects.

Introduction

Tinnitus is a frequent and often devastating symptom of disorders of the auditory system and a wide variety of other pathological conditions (Lechtenberg and Shulman, 1984, Seidman and Jacobson, 1996, Coles, 1997). The lack of an evident external sound stimulus associated with the experience of tinnitus has led to the definition of tinnitus as an auditory ‘phantom’ perception (Jastreboff, 1990). Many dysfunctions of the auditory system, resulting in aberrant neural activity, have been proposed as the cause of tinnitus (Jastreboff, 1990), yet no theory of the underlying pathophysiology has been substantiated. Most hypotheses claim that tinnitus is related to cochlear dysfunction (Coles, 1997). Damage to hair cells due to noise or ototoxic drugs, stereocilia decoupling, changes in calcium ion concentration or disturbance of the synaptic transmission were suggested most frequently as possible cochlear causes (Romand, 1992, Zenner and Ernst, 1993, Lenarz et al., 1995, Lepage, 1995). Pathologic changes in the acoustic nerve with cross-talk between adjacent fibers as a result of insufficient insulation (due, for example, to retro cochlear pathology) (Møller, 1984) or imbalance between the activity of large fibers innervating the outer hair cells and small inner hair cell fibers are other possible patho-mechanisms (Tonndorf, 1987, Møller, 1997). Additionally, abnormal activity at higher levels of the auditory pathways (auditory nuclei, auditory cortex, associative cortices) may contribute to the generation of tinnitus (Jastreboff et al., 1988, Meikle, 1995, Gerken, 1996, Jastreboff, 1996a). Despite this profusion of assumed locations of the generator of tinnitus, most current hypotheses agree that abnormal neural activity is interpreted and perceived as tinnitus in higher cortical centers (e.g. auditory cortex) (Jastreboff, 1990).

The interpretation of an aberrant auditory signal as ‘troublesome tinnitus’ not only implies a process of conscious sound processing in auditory centers, but must at the same time associate the signal with unpleasantness and distress. Memory, attention and the emotional state of the patients are factors supposed to be involved in this reaction. Thus, an evaluation of the cerebral function in tinnitus sufferers may uncover the underlying neurophysiological process.

The only clinically available measure of tinnitus is the psycho-acoustical description of pitch and loudness, which is based on a subjective match between tinnitus and external sounds (Goldstein and Shulman, 1981, Coles et al., 1984, Hallam et al., 1985). A satisfying match is often impossible. In other cases, it is unsatisfactory due to temporal fluctuations and poorly reproducible measurements. The match to the audiologic features gives some information about the characteristics of the sensation, but does not allow for predictions of the treatment outcome, or severity, or the degree of annoyance and dislike involved (Hazell et al., 1985, Jastreboff et al., 1994). Without useful objective measurements, confirmation and validation of tinnitus for medico legal purposes is impossible.

In an attempt to demonstrate objective evidence of tinnitus, a link to the presence of spontaneous otoacoustic emissions (SOAE) and tinnitus was suggested, but appeared to be weak (Wilson, 1986, Penner and Burns, 1987, Penner, 1990, Ceranic et al., 1995). Compared to controls with normal hearing, brainstem auditory-evoked responses in patients with tinnitus showed consistently significant differences (Shulman and Seitz, 1981, Ikner and Hassen, 1990, Lemaire and Beutter, 1995, Attias et al., 1996). Unfortunately, these differences could not be reproduced in other studies (Barnea et al., 1990, Møller et al., 1992). Conflicting results were also found in experiments using magneto-encephalography. Some studies reported that auditory-evoked magnetic fields in tinnitus sufferers were different from those of normally hearing individuals (Hoke et al., 1989, Pantev et al., 1989, Shiomi et al., 1997). Others failed to replicate these results (Jacobson et al., 1991, Colding-Jørgensen et al., 1992).

With the development of functional brain imaging techniques, such as single photon emission-computed tomography (SPECT), functional magnetic resonance imaging (fMRI), and positron emission tomography (PET), new potential methods for the objective measurement of tinnitus have emerged. These imaging techniques can reveal changes of activity in the central nervous system by measuring the regional cerebral blood flow (rCBF). Only a few studies have yet applied these techniques to tinnitus to reveal such changes (Shulman et al., 1995, Cacace et al., 1996a, Arnold et al., 1996, Lockwood et al., 1998). The body of results from those studies suggests the existence of an abnormal connection between the auditory cortex and the limbic system. It is hypothesized that the auditory system mediating the tinnitus sensation may activate emotion control systems and memory systems in the hippocampus. Thus, the involvement of such brain regions may explain the distress and annoyance associated with many cases of tinnitus.

We studied the functional neuroanatomy of the generation, perception and processing of tinnitus with PET in a group of patients with disabling tinnitus. Our goal was to test the working hypothesis that effective masking and suppression of the tinnitus sensation with lidocaine is associated with a decreased activity in one or more cortical regions involved in the perception of the tinnitus stimulus. PET images of rCBF distribution were obtained while the subjects experienced their habitual tinnitus and while this sensation was suppressed or removed. By adopting the subtractive approach on these PET data sets, we expected to isolate activation in specific neuroanatomical systems concerned with the cerebral representation of tinnitus. Based on prior work and theoretical considerations, we predicted that the sensation of tinnitus would recruit the primary and associative auditory cortices in the temporal lobe, the prefrontal cortex and the limbic system. Thus, we anticipated to find further support for theories of tinnitogenesis.