Threshold estimation in adult normal- and impaired-hearing subjects using auditory steady-state responses

Objective: To compare the estimation of hearing threshold values from behavioral and electrophysiological (ASSR) methods, in subjects with normal hearing and sensorineural hearing impairment. In particular we were interested in estimating : (a) the error margin of the ASSR estimated threshold levels with a commercial instrument (Audera) ; and (b) how the ASSR estimated hearing levels depend on the degree of hearing loss. Methods: We have tested 32 subjects (17 male and 15 female) for a total of 61 ears. From those 11 (22 ears) presented normal hearing threshold values ( 0- 19 dB HL) and 21 (39 ears) sensorineural deficits. The latter group was subdivided in three classes namely : (i) 11 subjects (16 ears) with moderate hearing loss (53.7 dB HL ±12.3); (ii) 5 subjects (11 ears) with severe hearing loss (80.6 dB HL ±12.7); and (iii) 6 subjects (12 ears) with profound hearing loss (101 dB HL ± 5.5). Results: The data show that for the normal hearing subjects the ASSR threshold is approximately around 20 dB ( 11 dB SD) for the frequencies 0.25 – 1.0 kHz. For the higher frequencies the ASSR threshold increases up to 40 dB (12.5 dB SD) at 8.0 kHz. In our hearing impaired subjects also, we observed this phenomenon. In fact the ASSR proved to reliably predict the behavioral threshold (+/- 5 dB) especially in the group or most impaired ears. Similar errors were found in the less impaired ears, and for the ASSR at high frequencies. The regression analysis confirmed that the difference between the ASSR-estimated and behavioral threshold values, significantly decreases with the amount of hearing loss. Our data evidenced that for a 10 dB increment of the behavioral threshold, the ASSR threshold increases of 7 dB. The difference between the two methods, of about 27 dB observed in normal subjects tends to cancel in the hearing loss greater then of 95-100 dB HL. Conclusion: The results of our analyses indicate that the threshold estimates are rather discordant with the behavioral thresholds. Particularly, it seems that the correction factor we have applied does not rely on factors adequately modeled (instrumentation-wise) to compensate for the effects of hearing loss on ASSR thresholds. The threshold estimation is adequately modeled for high levels of hearing loss particularly for patients needing a cochlear implant.

Background: The aim of the study was to compare the estimation of hearing threshold values by behavioral and electrophysiological (ASSR) methods in subjects with normal hearing and those with sensorineural hearing impairment. Material/Methods: Thirty-two subjects (17 male and 15 female) were tested, with a total of 61 ears. Of these, 11 (22 ears) presented normal hearing threshold values (0-19 dBHL) and 21 (39 ears) sensorineural deficits. Results: The data showed that for the normal-hearing subjects the mean ±SD ASSR threshold was approximately 20±11 dB for frequencies of 0.25-1.0 kHz. For higher frequencies the ASSR threshold increased to 40±12.5 dB at 8.0 kHz. Regression analysis confirmed that the difference between the ASSR-estimated and behavioral threshold values decreased significantly with the amount of hearing loss. The data showed that for a 10-dB increment of the behavioral threshold, the ASSR threshold increased by 7 dB. The difference of about 27 dB between the two methods observed in normal subjects tends to cancel in hearing loss greater than 95-100 dBHL. Conclusions: The results of these analyses indicate that the threshold estimates are rather discordant with the behavioral thresholds. in particular it seems that the correction factor applied here does not rely on factors adequately modeled (in terms of instrumentation) to compensate for the effects of hearing loss on ASSR thresholds. The threshold estimation is adequately modeled for high levels of hearing loss, particularly for patients requiring a cochlear implant. © Med Sci Monit, 2010.