The relationship between auditory threshold and evoked otoacoustic emissions

The aim of this study was to compare transiently evoked otoacoustic emissions (TEOAE) and distortion product otoacoustic emissions (DPOAE) in normal hearing ears (n = 44) and ears with cochlear hearing loss (HL) to obtain defined data on qualitative and quantitative correlations. In addition, we wanted to determine the reliability with which a clinical examiner could predict a typical, idealized audiometric configuration from TEOAE measurements. In the hearing-impaired subjects (n = 149), a 50% reduction of OAE incidence was caused by a mean HL of 10.5 dB for TEOAE compared to 27 dB SPL for DPOAE. A 90% incidence reduction was found at a mean threshold elevation of 33 dB for TEOAE and 51 dB for DPOAE. Correlation between TEOAE amplitudes and HL was in general rather low (r = –0.1 to –0.5), while DPOAE amplitudes showed a slightly better correlation with HL (r = –0.3 to –0.6). In general, efforts to derive an audiogram from evoked OAE have been more promising for DPOAE than for TEOAE. However, our studies showed that approximately 40% of the ears with HL could be categorized correctly into one of five typical audiometric patterns from TEOAE measurements. Additionally, a cochlear HL in or near the medium frequency range was much more likely to cause a reduction in TEOAE than an isolated low- or high-frequency lesion. Accordingly, TEOAE were often preserved in ears with isolated HL in the high or low frequencies. Received: 7 May 1998 / Accepted: 25 September 1998