Otitis media with effusion (known as 'glue ear' when it last >6 weeks) is common in children; it usually clears spontaneously, but if persistent it can have adverse effects on their learning and development and may require surgical intervention. There are no proven treatments effective in primary care, however some evidence suggests a benefit from nasal topical corticosteroid: this trial aimed to test the efficacy of this treatment in a primary care setting. Participants were children aged 4 to 11 years with a recent (<12 months) history of otitis media or a similar ear problem, and presence of confirmed bilateral otitis media with effusion. They were randomised to three months treatment with mometasone nasal spray or matching placebo, used once daily into each nostril. Symptoms were recorded in diaries provided. The primary outcome was the proportion cured in at least one ear by objective assessment (tympanometry) at one month; other outcomes included cure rates at three and nine months, adverse effects, and diary symptoms.

Original power calculation set a target of 240 children, however the trial was terminated after 217 had been randomised because recruitment was slower than expected: independent statistical advice indicated that by that point, there was no chance that the main findings would be reversed. There were 105 children in the active corticosteroid group and 112 in the placebo group: 9 and 14 respectively were lost to follow-up or had missing / un-interpretable tympanometric data before the one month assessment, which was therefore based on 96 children receiving corticosteroid and 98 receiving placebo.

There was no significant difference between the groups for the primary endpoint: 41% (39/96) of the topical steroid group and 45% (44/98) of the placebo group were cured in one or both ears at one month (difference favouring placebo 4.3%; 95% CI –9.3% to 18.1%). There were also no significant differences in the secondary outcomes studied, including adverse effects and diary symptoms.

The authors conclude that nasal topical corticosteroid treatment does not improve outcomes in children with bilateral otitis media with effusion treated in primary care. This therapy is therefore unlikely to be useful in such children. They note that there was a high rate of natural resolution, and discuss possible reasons for the negative result.