Sub-classification of tinnitus with and without hyperacusis by combining audiometry, gamma oscillations, and hemodynamics

The neuronal basis of tinnitus is still not fully understood, which delays the development of causal therapies. The aim of the present study was to identify objective characteristics of tinnitus without (T) or with concomitant hyperacusis (TH). For this purpose, we performed electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS) while subjects had to perform an auditory discrimination task within their tinnitus frequency and a reference frequency as well as functional MRI examination (resting-state fMRI-based functional connectivity (rs-fMRI-bfc)). The T group showed prolonged brainstem audiometry (ABR) latency and decreased ABR-V amplitudes, which was associated with lower rs-fMRI-bfc between ascending auditory nuclei and primary auditory cortex, as observed in previous studies. In the T subjects, these features were associated with increased spontaneous and reduced evoked gamma oscillations and with reduced deoxy-Hb concentrations in response to lower frequency stimuli in temporal cortex (Brodmann area (BA) 41, 42, 22), suggesting less synchronous auditory responses during active auditory discrimination of reference frequencies. In contrast, in the TH group, gamma oscillations and hemodynamic responses in the temporoparietal regions were reversed during active discrimination of tinnitus frequencies. In conclusion, we were able to show that tinnitus without and with hyperacusis can be distinguished by the EEG and fNIRS paradigms used, which may allow a more precise subclassification of tinnitus and future improved treatment approaches.