Decongestant-antihistamine preparations used alone or in combination with an antimicrobial agent have failed to improve the course of otitis media with effusion. Studies using short-course steroid therapy have varied widely in method and shown inconsistent results. In this study, Mandel and colleagues investigate steroids and amoxicillin in the treatment of middle ear effusion.

The authors define middle ear effusion as either otitis media with effusion or acute otitis media. Patients one to nine years of age with different durations of effusion (but no less than two months, except in cases of unknown duration) were stratified to one of four treatment arms: (1) steroid with amoxicillin for 14 days, then amoxicillin for 14 more days; (2) steroid with amoxicillin for 14 days, then amoxicillin placebo for 14 days; (3) steroid placebo and amoxicillin for 14 days, then amoxicillin for 14 more days; and (4) steroid placebo and amoxicillin for 14 days, then amoxicillin placebo for 14 more days. The children were examined at two and four weeks and were evaluated using acoustic immittance and pneumatic otoscopy based on a previously designed algorithm. Allergy and hearing tests also were performed.

At two weeks, 33.3 percent of children receiving steroids plus amoxicillin had no middle ear effusion compared with 16.7 percent in the placebo and amoxicillin group ( P = 0.03). At four weeks, 32.8 percent of the steroid group had no middle ear effusion compared with 20 percent in the placebo group ( P = 0.12). The authors also analyzed the change in middle ear effusion status from the two-week visit to the four-week visit. Of children with no middle ear effusion at two weeks, 47.8 percent in the steroid-treated group had recurrence of effusion, as did 45.5 percent of those receiving placebo ( P = 0.69). Of the children who still had middle ear effusion at two weeks, 20.9 percent of the steroid-treated group had no middle ear effusion by four weeks, as opposed to 13 percent in the placebo group ( P = 0.31). In addition, 36.8 percent of children on the longer course of amoxicillin had recurrence of middle ear effusion by four weeks, as opposed to 60 percent of those taking placebo ( P = 0.17). Of those with no middle ear effusion at two weeks, 17.4 percent cleared with the longer amoxicillin course, whereas 15.7 percent cleared with placebo ( P = 0.91).

Hearing in the steroid group improved significantly at two weeks, but only specific aspects of hearing remained significantly improved after four weeks. There were no statistically significant differences in hearing at the four-week visit among those who had received amoxicillin or placebo for the second two-week period.

The authors found a significant difference in the proportion of children who were effusion-free immediately after two weeks of steroid (prednisolone) and amoxicillin as opposed to those treated with amoxicillin only. However, this difference disappeared during the two weeks following completion of treatment. The authors conclude that the dosage and type of steroid used in this study should not be used routinely in the treatment of otitis media with effusion.