A new approach to serial monitoring of ultra-high frequency hearing.

Early detection of drug-induced hearing loss is best accomplished by monitoring hearing at the ultra-high frequencies. Unfortunately, at these frequencies, sound pressure at the tympanic membrane (TM) critically depends on the placement of the sound source and on the size and shape of each individual external ear. Thus, presentation of the same sound may yield substantially different sound pressures in different ears. Moreover, only a slight change in the position of an earphone may yield large changes in sound pressure at the tympanic membrane. As a consequence of these characteristics, the reliability of ultra-high-frequency audiometry is poorer than at conventional audiometric frequencies. However, for the early detection of ototoxicity, it is necessary only to monitor for increases in thresholds. Accordingly, a sound-delivery system was developed which fixes the relative position of the sound source and the ear. This system ensures that sounds at the same level may be presented during different test sessions. To assess the stability of ultra-high-frequency thresholds, normal hearing subjects were tested in sessions separated by several weeks. Thresholds were obtained between 1 and 16 kHz and were found to be reliable. It is concluded that this type of system can be used for monitoring the ototoxic effects of drug therapy.