Letters to the Editor Raised ABR threshold after suction aspiration of glue from the middle ear: three case studies Dear Sir, I have read with interest the paper by Mason et al. in the August issue of the Journal. Auditory brainstem response (ABR) thresholds measured immediately after suction aspiration of middle ear fluid were compared with hearing assessment, in some cases with repeat brainstem response audiometry (BRA), at a later date. In six of 14 ears the latter threshold was improved by 15 dB or more. These results are interpreted as indicating a possible temporary threshold shift due to suction noise. There is, however, another possible explanation. We have previously shown that the time-course of hearing threshold improvement following myringotomy, aspiration of middle ear secretion and insertion of a ventilation tube is frequency dependent (Mair etal, 1989). Threshold improvement in the low frequency range of 0.25-1 kHz is immediate, is delayed at 4-8 kHz and further delayed, by two to eight weeks, in the extra-high frequency range of 9-20 kHz. Since click-evoked ABR thresholds at moderate to high intensities originate from the 2-8 kHz region of the cochlea (Eggermont and Don, 1980; Burkard and Hecox, 1983), a persistent threshold elevation at 4 and 8 kHz immediately after myringotomy, suction and grommet insertion would result in a high ABR theshold. The use of 1 kHz pure-tone pips as ABR stimulus, especially when high-pass masking is not employed, results in high thresholds even in normal-hearing adults, a correction factor of 30-40 dB having been indicated (Fjermedal and Laukli, 1989). The threshold differences reported by Mason et al. (1995) may be due to the frequency dependent timecourse of threshold improvement following tubulation and, possibly, the inherent difficulties associated with response identification with low-frequency stimuli.