gms | German Medical Science

Norena et al. [Ref. 1] proposed a method for the psychoacoustic characterization of tinnitus. The aim of our study was to investigate clinical uses of psychoacoustic tinnitus spectra as well as their relation to cochlea defects.

Methods: 61 tinnitus patients (28 with normal hearing, 33 with high frequency hearing loss) treated at the University of Saarland clinic were enrolled in the study. They took a complete audiological examination as well as the TEN Test described by Moore [Ref. 2] and psychoacoustic tinnitus spectra (Norena et al., 2002).

Psychoacoustic tinnitus spectra were measured in a similar way as proposed by Norena et al. [Ref. 1]. The standard audiometric test frequencies were presented three times in a randomized order. The sound level was increased until the patient signaled that the sound was as loud as his own tinnitus. Then the patient had to rate the resemblance with his own tinnitus on a visual analog scale from 0 to 10. Tinnitus spectra were calculated as the mean rating at each frequency.

Dead regions of the cochlea (regions without functioning inner hair cells) were identified using the TEN Test.

Results: The psychoacoustic tinnitus spectra could be classified into different types:

Most of the patients presented with a tinnitus spectrum with a nearly linear characteristic. This results from a tonal high pitched tinnitus.

In other cases tinnitus spectra showed multiple local maxima. This could be the result of different distinct tinnitus components which the patient experiences as a chord like sound.

Some tinnitus spectra showed high resemblance levels at neighbouring frequencies which could be the characteristic for a narrow band noise.

Comparing the mean of all audiograms with the mean of all tinnitus spectra a clear correlation between hearing loss and tinnitus spectra could be observed.

Dead regions could be shown in 82% of patients with tinnitus and high frequency hearing loss using the TEN Test. Patients with high frequency hearing loss but without tinnitus showed dead regions in 55%. While the main peak of the tinnitus spectra were often not associated with the edge of a hearing loss, 85% of the cases with dead regions showed a close connection between the most dominant peak of the tinnitus spectrum and dead regions.

Conclusion: Psychoacoustic characterized tinnitus spectra give a more detailed representation of a patient’s tinnitus as compared to the pitch identification using an alternative choice method, especially in cases where more than one spectral component is present.

Practical uses could be tinnitus counseling and studies which require a more precise reconstruction of the subjectively experienced Tinnitus.

Psychoacoustic measured tinnitus spectra combined with the results of the TEN Test show that in many cases not the edge of the hearing loss but the edge of a dead region could have a pathogenetic influence on tinnitus.