Features predicting treatment failure in pediatric acute otitis media

To facilitate better antibiotic stewardship, we conducted this clinical trial to identify the prognostic features of treatment failure in pediatric acute otitis media (AOM).

Design: This is a randomized, parallel-group, open-label, comparative clinical trial.

Children with AOM and aged between 1 month and 5 years were enrolled. Patients were randomly assigned to receive either amoxicillin alone (70 mg/kg) for five days, or the same with additional clarithromycin (15 mg/kg) for the initial three days. The clinical course of AOM was evaluated based on tympanic membrane scores. Failure of treatment for AOM was confirmed on day 14. Nasal conditions were also assessed by a clinical scoring system for acute rhinosinusitis.

Treatment failures occurred in 25 out of 129 (19.4%) children. The ratio of treatment failures by age was significantly higher in children younger than 2 years than in children older than 2 years. The tympanic membrane scores on day 3 (P = 0.0334) and day 5 (P < 0.0001) and acute rhinosinusitis scores on day 5 (P = 0.0004) were higher in failure cases than in cured cases. Multivariate logistic regression analysis indicated significant associations between the treatment failure with tympanic membrane scores and acute rhinosinusitis scores on day 5, and the antimicrobial treatment regimen.

Improvement of acute rhinosinusitis and tympanic membrane scores on day five were important predictive features in failure of treatment for pediatric AOM. These results will be useful when discussing the treatment decisions with the patient's parents.