A meta-analysis of individual patient data from 6 RCTs of the effects of antibiotics in children aged 6 months to 12 years with acute otitis media (AOM) included 1643 subjects 1. The overall rate difference in extended otitis media (the primary outcome), defined as pain, fever, or both at 3-7 days, was -13%, 95% CI -9% to -17%, NNT = 8. In children younger than 2 years of age with bilateral AOM, 55% of controls and 30% on antibiotics still had extended otitis, with a rate difference between these groups of -25% (95% CI -36% to -14%), NNT = 4. There were no significant differences for age alone. In children with otorrhoea the rate difference was -36% (-53% to -19%), NNT = 3, whereas in children without otorrhoea the equivalent values were -14% (-23% to -5%), NNT = 8. None of the children in the trials developed mastoiditis.
Comment: The authors state that the prespecified subgroups were selected on the basis of an unpublished earlier study within the same setting. The quality of evidence is downgraded by indirectness. Pain and fever are not the only important outcomes; the duration of effusion and hearing impairment should also be considered. The diagnostic accuracy in the includes studies has also been questioned.
A recent American Academy of Pediatrics (AAP) guideline 2 recommends that children aged 6 months to 2 years with AOM to be treated with antimicrobial treatment. However, they suggest observation without initial antimicrobial therapy for children in whom AOM is unilateral and unaccompanied by severe signs or symptoms.
Primary/Secondary Keywords