Abstract:
Tinnitus is a common condition caused by a dysfunction of the auditory system. The
development of effective therapeutic interventions for tinnitus is challenging due to
debate regarding its causative neural mechanisms. In Europe, approximately one in seven
adults (14.7 %) report tinnitus symptoms (Biswas et al., 2022), resulting in significant
personal distress and substantial socioeconomic costs. This thesis aimed to address the
possible contribution of co-occurring hyperacusis as well as the tinnitus duration to the
underlying neural mechanisms of tinnitus.
Three comprehensive studies were conducted with a total of 128 subjects: 45 tinnitus
patients without hyperacusis (T-group), 26 tinnitus patients with hyperacusis (TH-group),
and a control group of 57 subjects. In order to ensure the reliability of our findings, we
adopted a multimodal approach to detect and exclude unwanted comorbidities. The
multidimensional approach integrated questionnaire evaluations (Goebel-Hiller-Score
and Hyperakusis-Inventar), audiological diagnostics (e.g., pure tone audiometry, speech
audiometry, uncomfortable loudness level, tinnitus localisation, and supra-threshold
auditory brainstem response (ABR)). Furthermore, I evaluated changes in neural activity
using cortical haemodynamic responses (functional near-infrared spectroscopy (fNIRS)
and blood oxygenation level-dependent (BOLD) functional magnetic resonance imaging
(fMRI)) and electric potentials (electroencephalography (EEG)) in anatomically predefined
brain regions.
The results distinguish or allow for differentiation between patients with recent onset of
tinnitus and those with chronic tinnitus: TH patients reported higher levels of distress and
annoyance from the beginning, with an increased tinnitus loudness in patients with long
term tinnitus, while in the T-group, the tinnitus loudness declined over time. In addition,
the T-group experienced a shift from primarily unilateral (83 %) to utterly bilateral tinnitus
percept, whereas the TH-group had bilateral tinnitus from early on (75 % bilateral).
Moreover, the findings revealed distinct patterns emerging among tinnitus patients with
and without hyperacusis, both at rest and in response to acoustic stimuli. Although no
differences were observed in pure tone audiometry, the hearing threshold in the range of
the individual tinnitus frequency correlated positively with the tinnitus loudness and
distress in the T-group, whereas the TH-group experienced maximal distress levels
already at minimal tinnitus loudness. The T-group exhibited delayed and reduced ABR
wave responses, diminished frequency of positive resting-state correlations and reduced
evoked BOLD fMRI responses of auditory-associated regions, particularly in response to
high-frequency stimuli. Conversely, the TH-group was distinguished based on less
reduced ABR wave V responses, less reduced frequency of positive resting-state
correlations in ascending (sub-) cortical auditory-associated regions, and enhanced BOLD
fMRI responses of the thalamus, auditory cortex, hippocampus, and posterior insula,
particularly in response to low-frequency stimuli. In addition to the group differences in
slow cortical haemodynamic responses described above, the T-group exhibited increased
spontaneous and reduced 1 kHz stimulus-evoked fast gamma-band oscillations in the
Auditory Cortex and Broca's area. In contrast, the TH-group was characterised by
distinctions in the left temporo-parietal and dorsolateral PFC, potentially associated with
excessive auditory arousal spreading to attention and pain networks. In summary, these
results support the tinnitus theory of reduced auditory responsiveness, best explained by
hyperexcitability through the loss of tonic parvalbumin (PV) inhibitory interneurons in
deprived regions due to the critical loss of high spontaneous firing rate fibres in the Organ
of Corti. This interpretation would be consistent with the observed elevation in
spontaneous gamma activity within the auditory cortex of the T-group, which could be
attributed to the loss of tonic inhibition of pyramidal neurons through PV interneurons.
The present findings provide evidence for distinct neural correlates of tinnitus and tinnitus
with co-occurring hyperacusis, emphasising the need for a sub-entity-specific differential
diagnosis in therapy and research, as hyperacusis is one of the main causes of distress
and tinnitus complaints over time. Furthermore, the identified characteristic neuronal
profiles provide valuable functional biomarkers for the follow-up of therapeutic trials and
should guide medical tinnitus practice towards personalised curative therapies.