Evaluation of the Treatment of Aspiration Pneumonia in Hospitalized Children.

Background: Aspiration pneumonia (AP) treatment is variable with limited available guidance on optimal antibiotic choice and duration. This study evaluated the impact of antibiotic regimen and duration on treatment failure for AP in children and correlated the effects of antimicrobial stewardship program (ASP) interventions on treatment duration.
Methods: Hospitalized children who received antibiotics for AP were identified through an existing ASP repository. Diagnosis was confirmed through ASP documentation with either an international classification of diseases 9/10 code or physician diagnosis of AP. Incidence of treatment failure (necrotizing pneumonia, lung abscess, empyema, or retreatment) was compared between patients receiving shorter (≤7 days) vs longer (>7 days) course of antibiotics and between various empiric/final antibiotic regimens utilized. Duration of treatment was evaluated in patients with or without an ASP intervention.
Results: Four hundred and nineteen treatment courses for AP were included. Nineteen episodes (4.5%) of treatment failure were identified. No difference in treatment failure was observed between shorter vs longer courses (8 vs 11 episodes). An aminopenicillin plus beta-lactamase inhibitor was most frequently utilized for both empiric (47.2%) and final treatment (67.5%). Treatment failure rates did not differ with length of intravenous therapy nor empiric/final antibiotic regimen chosen. ASP interventions targeting duration were associated with significantly shorter courses (6.28 vs 7.46 days; P = .04).
Conclusions: Shorter courses of antibiotics did not result in more treatment failure for AP when compared to longer courses. Neither antibiotic choice nor route impacted treatment failure rates. ASPs may optimize the treatment of pediatric AP.
(© The Author(s) 2021. Published by Oxford University Press on behalf of The Journal of the Pediatric Infectious Diseases Society. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.)