Antibiotic treatment has been shown to speed resolution of the symptoms of acute otitis media in children, but since there is no significant difference in long-term outcomes among children who are treated and children who are not treated, some clinicians advocate not prescribing antibiotics at all. There is, however, a reluctance on the part of American physicians to adopt this practice. Kozyrskyj and associates conducted a meta-analysis to determine if a short course of antibiotic treatment was as efficacious as a traditional longer course of therapy for acute otitis media in children.

Clinical trials were included if their study populations were between four weeks and 18 years of age, had a diagnosis of acute otitis media, had been randomly assigned to receive a short or long course of antibiotic treatment and were assessed to see if the acute otitis media had resolved. Treatment failure was defined as lack of clinical resolution, or relapse, or recurrence within one month. Middle ear effusion was not considered a treatment failure.

The literature search yielded 32 trials: 17 were trials of short-acting antibiotics, 11 were trials of azithromycin and four were trials of ceftriaxone. Fourteen of these studies could be used for evaluation of the main outcome measures. A total of 1,031 children were treated for five days with an antibiotic, and 1,084 children were treated for eight to 10 days. There were no significant differences in primary outcomes between the two groups. Evaluation at 30 days after initiation of therapy showed that a 10-day course of antibiotics was not significantly different from a five-day course of antibiotics; however, at 20 days there was a slight increase in relapse or reinfection (odds ratio: 1.52) in those treated with a shorter course of antibiotics. The difference in short-term risk of failure was 7.8 percent. In other words, to prevent one treatment failure with the short course of therapy, 13 children would need to receive the longer course of antibiotics.

The authors conclude that a five-day course of antibiotic treatment for uncomplicated acute otitis media in children is effective, costs less and is associated with greater compliance than the longer course of therapy that is typically prescribed.

In a related editorial, Pichichero states that many children are overdiagnosed with acute otitis media if tympanocentesis is not performed. He stresses that diagnostic accuracy must be improved, particularly by tympanocentesis. Tympanocentesis should be used in patients whose tympanic membrane is bulging and in patients who appear toxic or have a high fever. Long (10-day) courses of antibiotic therapy should be reserved for use in patients who have perforated tympanic membranes, those with chronic or recurrent otitis media and those with underlying medical conditions who may be at higher risk of treatment failure.