Skip navigation

Steroids for glue ear


Clinical bottom line

There is little good evidence that steroids are beneficial, but in four trials with 274 children, hearing loss at two weeks was less with steroids than control, with an NNT of 3.7 (2.6 to 6.4)


CC Butler & JH van der Voort. Steroids for otitis media with effusion. A systematic review. Arch Pediatr Adolesc Med 2001 155: 641-647. [Note that more detailed information can be found in a version of the review in the Cochrane Library]


Randomised controlled trials of oral or intranasal steroids. Blinding had to be adequate. Outcomes were children not ears. No additional treatment other than steroids was allowed for included studies. Childrens' ages were up to 12 years. Significant hearing loss was defined by a 20 decibel loss twice or more within three months. Very wide searching strategies were used.


There were very few studies, with very few children, which made for difficulty in assessing the benefits of steroid therapy. The largest data set was four studies of steroids plus antibiotics versus antibiotics alone, with 274 children. Two week resolution data were available.

Hearing loss at two weeks was present in 66/141 (47%) children treated with steroids with antibiotics, and in 98/133 (74%) children treated with antibiotics alone (Figure 1). The relative benefit was 0.64 (95%CI 0.52 to 0.79) for hearing loss at two weeks. The number needed to treat was 3.7 (2.6 to 6.4).

Figure 1: Hearing loss at two weeks in children treated with steroids plus antibiotics or antibiotics alone (control)


The evidence for the benefits of steroids in children with acute otitis media with effusion (glue ear), used alone or in combination with other treatments, is imperfect.