Determining the Cause of Hearing Loss: Differential Diagnosis Using a Comparison of Audiometric and Otoacoustic Emission Responses

In Brief Objective: The purpose of this study was to further investigate the possibility of developing noninvasive methods of differential diagnosis of hearing disorders through the study of experimental animals with induced lesions. In particular, it was desired to compare distortion product otoacoustic emission (DPOAE) responses and auditory brain stem response (ABR) thresholds in Mongolian gerbils having either acoustic or strial damage, using as a reference the same responses measured in a control group of normal young adult gerbils. The goal was to evaluate the potential clinical application of this approach to determining the dominant contribution to sensorineural hearing loss in individual human subjects. Design: DPOAE input-output functions and ABR thresholds were measured over a wide range of stimulus frequencies for three groups: (1) a reference group of normal young adult gerbils; (2) a group in which acoustic damage had been induced 2 wk earlier; (3) a group in which damage to the stria vascularis was induced by a series of furosemide injections. The responses in the experimental groups relative to the normal means were compared to determine which combinations of responses were effective in discriminating between animals with different lesions. Three measures were evaluated in detail: the ABR threshold, the emission threshold at a criterion emission amplitude, and the emission amplitude at a high stimulus level. Results: Considering cases with significant hearing loss (ABR thresholds elevated by 20 dB or more), the best method for distinguishing between the two lesions involved a two-dimensional plot comparing emission and ABR thresholds at the same stimulus frequencies. Acoustic damage cases were found in a broad region where the emission and ABR thresholds were roughly equal, whereas strial damage cases were found in a narrower region where the emission threshold was about 0.4 times the ABR threshold (both in dB). These two cases were compared with a third case introduced by definition, that is, damage to inner hair cell or neural systems resulting in an increase in audiometric threshold but no change in emission responses (e.g., auditory neuropathy). The responses for these three cases were found to lie in different regions of the two-dimensional plot comparing emission and ABR thresholds, provided only that ABR thresholds were elevated 20 dB or more. This diagram also revealed cases of preclinical acoustic damage, in which the ABR threshold was shifted less than 20 dB but where the emission threshold was significantly elevated. Conclusions: The results clearly demonstrate the possibility of developing a clinical method of noninvasive differential diagnosis of hearing loss. The method demonstrated was to add to a standard audiometric evaluation the measurement of DPOAE growth functions over the range of frequencies where these emissions were relatively easy to measure and consistent. The DPOAE stimulus frequencies were chosen to match the audiometric frequencies, and the corresponding emission and audiometric thresholds were compared on a threshold-threshold plot for each individual at a number of stimulus frequencies. Responses in different regions in this plot were found to correspond to different types of sensorineural hearing loss. It is becoming increasingly important to have a non-invasive method of determining the cause of sensorineural hearing impairment, i.e., to determine if the main cause is dysfunction in outer hair cell, endocochlear potential, or inner hair cell/VIIIth nerve pathway. It is argued that the underlying cause cannot be uniquely determined from an audiogram alone. In this study, both distortion product otoacoustic emissions and auditory brain stem responses were measured in normal gerbils and in animals in which different types of lesions had been introduced. It was found that the best method for distinguishing the type of lesion involved a direct comparison of emission and auditory thresholds. For outer hair cell damage, responses were found in a broad region where the increase in emission threshold was about equal to that of the auditory brain stem response. For endocochlear potential dysfunction, responses were found in a narrow region where the emission threshold was about 0.4 of the auditory threshold. These responses were non-overlapping, so that a comparison of emission and auditory thresholds successfully provides a method for clinical diagnosis of the cause of sensorineural hearing loss.