Equivalence and test–retest reproducibility of conventional and extended-high-frequency audiometric thresholds obtained using pure-tone and narrow-band-noise stimuli

Abstract

Objectives: Extended high-frequency (EHF) audiometry is useful for evaluating ototoxic exposures and may relate to speech recognition, localisation and hearing aid benefit. There is a need to determine whether common clinical practice for EHF audiometry using tone and noise stimuli is reliable. We evaluated equivalence and compared test-retest (TRT) reproducibility for audiometric thresholds obtained using pure tones and narrowband noise (NBN) from 0.25 to 16 kHz. Design: Thresholds and test-retest reproducibility for stimuli in the conventional (0.25–6 kHz) and EHF (8–16 kHz) frequency ranges were compared in a repeated-measures design. Study sample: A total of 70 ears of adults with normal hearing. Results: Thresholds obtained using NBN were significantly lower than thresholds obtained using pure tones from 0.5 to 16 kHz, but not 0.25 kHz. Good TRT reproducibility (within 2 dB) was observed for both stimuli at all frequencies. Responses at the lower limit of the presentation range for NBN centred at 14 and 16 kHz suggest unreliability for NBN as a threshold stimulus at these frequencies. Conclusion: Thresholds in the conventional and EHF ranges showed good test-retest reproducibility, but differed between stimulus types. Care should be taken when comparing pure-tone thresholds with NBN thresholds especially at these frequencies.