Background: The objective of the study was to evaluate hearing threshold errors induced by the use of a new acoustical protocol (Cochlea-Scan).
Material/methods: Thirty-five normally hearing and 30 hearing-impaired adults were used in the study. Hearing threshold levels were assessed by pure tone audiometry (PTA) and the Cochlea-Scan (Fischer-Zoth/Natus) DPOAE protocols.
Results: In the normal-hearing group, the mean difference in hearing levels at 2.0, 3.0, and 4.0 kHz was below 8 dB HL. At 1.5 kHz the mean difference was around 9 dB HL, while at 6.0 kHz the behavioral threshold was overestimated by 2 dB HL. Although linear regression models suggested a significant relationship between the behavioral and the Cochlea-Scan data in all the tested frequencies, the low R(2) values in each model suggest an unexplained variability in the data. For the hearing-impaired group the mean difference in hearing levels at 1.5 and 2.0 kHz was below 5 dB, while at 4.0 kHz the behavioral threshold was overestimated by 7 dB. In this group the observed relationships were stronger (larger R(2) values) than in the normal group, and the most significant relationships were observed at 2.0, 4.0, and 6.0 kHz (with descending R(2) values).
Conclusions: The data from this pilot study indicate that Cochlea-Scan assessment is possible and accurate in both normal-hearing and in hearing-impaired subjects. The margin of assessment in the hearing-impaired group was found to be closer to a PTA value of 40 dB and not 50 dB HL as previously suggested.