Objective: To evaluate the role of otoacoustic emission in children with middle-ear effusion and grommets.

Materials and methods: A prospective study was carried out on a total of 90 ears. All children listed for grommet insertion had a pre-operative and post-operative (three to six months after grommet insertion) pure tone audiometry, tympanometry and otoacoustic emission recorded. A comparison was made between pure tone audiometry and otoacoustic emission both pre-operatively and post-operatively.

Results: Pre-operatively, 63 ears had an abnormal pure tone audiometry of which 59 had absent otoacoustic emission. Therefore the sensitivity of otoacoustic emission in detecting a conductive loss was 59/63 = 94 per cent (95 per cent confidence interval 85 to 98 per cent). All 27 ears with normal hearing pre-operatively had normal otoacoustic emission. The specificity of otoacoustic emission was 27/27 = 100 per cent, (95 per cent confidence interval, 88 to 100 per cent). The positive predictive value was 59/59 = 100 per cent (95 per cent confidence interval, 94 to 100 per cent). After three to six months all post-operative patients with grommets had a normal pure tone audiometry and otoacoustic emission. So both pure tone audiometry and otoacoustic emission were strongly related both in patients with middle-ear effusion and in patients with grommets.

Conclusion: As the demonstration of hearing in young and difficult-to-test children can be problematic and time-consuming, we suggest that otoacoustic emission can be used as an alternative to pure tone audiometry in patients with middle-ear effusion and grommets.