We aimed to investigate whether auditory brainstem response (ABR), assessed immediately after grommet insertion, was an accurate measure of hearing threshold.

Ten children were recruited (20 ears). Patients underwent transient evoked otoacoustic emission (TEOAE) and ABR tests before and after grommet insertion under general anaesthetic. A third assessment was conducted at a six-week review, without anaesthetic or sedation. The hearing threshold in dB and the presence of TEOAE were recorded by senior audiologists. The results of each investigation were compared. At review, only 15 ABR and 17 TEOAE tests, of a possible 20, were recorded.

Transient evoked otoacoustic emissions were recorded in 30 per cent of ears pre-operatively, 20 per cent of ears post-operatively and 82 per cent of ears at six-week review. The average ABR thresholds were 23.75 dB (range 10–45 dB) pre-operatively, 27.25 (range 15–40 dB) post-operatively and 15 dB (range 5–25 dB) at six-week review. Comparing pre- and post-operative assessment, decreased thresholds were seen in six ears and elevated thresholds in 11. Comparing the six-week review ABR with the post-operative ABR, 13 ears had decreased thresholds, by as much as 20 dB. Thresholds of 35 dB or greater were measured in six ears immediately after grommet insertion (maximum 40 dB). Threshold shifts of 15 dB or greater was observed in seven ears. The intra-subject variation was shown to be significant (p<0.001) when repeated measures analysis of variance was applied.

We conclude that there is evidence that ABR can be inaccurate after grommet insertion, overestimating hearing thresholds. We believe that the mechanism by which the error is realised is temporary threshold shift.

In this study 22 patients (44 ears) with noise-induced permanent hearing loss were audiologically evaluated using transient-evoked otoacoustic emissions (TEOAE) and auditory brain-stem response (ABR). Twenty-one normal subjects (42 ears) without exposure to occupational noise were used as controls. Based upon the hearing loss at 4, 3, 2 and 1 kHz on the pure-tone audiogram, they were classified into four groups. In group 1 (eight ears), emissions were present in all ears but their TEOAE-noise level and their reproducibility (percentage) proved to be weak. The auditory brain-stem response (ABR) indicated that the I/V amplitude ratio, the latency values of wave V and the I–V intervals fell within the normal range in all ears. In Group 2 (14 ears), 40 per cent had no emissions, whereas the remaining ears showed weak emissions. The ABR revealed that in all ears the I/V amplitude ratio became small while wave V peak latency as well as I–V intervals were within the normal range. In Group 3 (10 ears), emissions were absent in 50 per cent, while in the other ears the emissions were very weak. The ABR revealed that the I/V amplitude ratio, which could be calculated in the 60 per cent in which wave I was present, was smaller than in Group 2. Wave V latency as well as I–V intervals were within the normal range. In Group 4 (12 ears), none of the ears showed emissions. The ABR indicated that the I/V amplitude ratio was much smaller when wave I was present (27 per cent) as well as I–V interval values being within the normal range. Wave V absolute latency value (δV index) indicated a positive index in 17 per cent of this group (two ears) when wave I was absent. In the present study a dynamic process from cochlear outer hair cells to cochlear neurons was seen, correlating with an increasing hearing loss.