Correlations between audiogram and objective hearing tests in sensorineural hearing loss.

Owing to its subjective nature, behavioral pure-tone audiometry often is an unreliable testing method in uncooperative subjects, and assessing the true hearing threshold becomes difficult. In such cases, objective tests are used for hearing-threshold determination (i.e., auditory brainstem evoked potentials [ABEP] and frequency-specific auditory evoked potentials: slow negative response at 10 msec [SN-10]). The purpose of this study was to evaluate the correlation between pure-tone audiogram shape and the predictive accuracy of SN-10 and ABEP in normal controls and in patients suffering from sensorineural hearing loss (SNHL). One-hundred-and-fifty subjects aged 15 to 70, some with normal hearing and the remainder with SNHL, were tested prospectively in a double-blind design. The battery of tests included pure-tone audiometry (air and bone conduction), speech reception threshold, ABEP, and SN-10. Patients with SNHL were divided into four categories according to audiogram shape (i.e., flat, ascending, descending, and all other shapes). The results showed that ABEP predicts behavioral thresholds at 3 kHz and 4 kHz in cases of high-frequency hearing loss. Also demonstrated was that ABEP threshold estimation at 3 kHz was not affected significantly by audiogram contour. A good correlation was observed between SN-10 and psychoacoustic thresholds at 1 kHz, the only exception being the group of subjects with ascending audiogram, in which SN-10 overestimated the hearing threshold.