OBJECTIVES: To evaluate the association between prior community-based antibiotic treatment for acute otitis media (AOM) and the incidence and severity of complications in children hospitalized with acute mastoiditis (AM), noting that this design does not address whether prior antibiotic treatment prevents AM development. METHODS: This retrospective cohort study analyzed medical records of 522 children (age 48 h before hospitalization versus those receiving < 48 h or none. The primary outcome was the presence of complicated AM, defined as extracranial (subperiosteal abscess) or intracranial (epidural abscess, sigmoid sinus thrombosis, brain abscess, or meningitis) complications. RESULTS: Of 522 children, 110 (21%) received prior antibiotics and 412 (79%) did not. While the composite complication rate was higher in the prior treatment group (54.5% vs. 38.6%, unadjusted OR = 1.91, p = 0.003), prior antibiotic treatment was not significantly associated with specific complication types, including subperiosteal abscess (adjusted OR = 0.67, p = 0.37) or intracranial complications (adjusted OR = 1.74, p = 0.13). Inflammatory markers, hospitalization length, imaging use, and surgical intervention rates were comparable between groups. Fusobacterium necrophorum in blood cultures was significantly associated with complications regardless of prior antibiotic exposure. CONCLUSIONS: Among children hospitalized with AM, prior antibiotic treatment was not associated with increased rates of specific complications. The difference in composite outcomes likely reflects selection bias rather than a true treatment effect. Causal inference is limited by confounding by indication and unmeasured baseline severity differences.
Sofer et al. (2026) studied this question.