Cargando…

Comparative Effectiveness of β-lactams Vs Azithromycin for Treatment of Outpatient Pediatric Community-acquired Pneumonia

BACKGROUND: Guidelines recommend amoxicillin as first-line therapy for mild, community-acquired pneumonia (CAP) in healthy, immunized children because of its effectiveness against S. pneumoniae. However, macrolides, which have inferior anti-pneumococcal activity, are the most commonly prescribed cla...

Descripción completa

Detalles Bibliográficos
Autores principales: Handy, Lori, Gerber, Jeffrey S, Bryan, Matthew, Zaoutis, Theoklis, Feemster, Kristen
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Oxford University Press 2017
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5630985/
http://dx.doi.org/10.1093/ofid/ofx162.007
Descripción
Sumario:BACKGROUND: Guidelines recommend amoxicillin as first-line therapy for mild, community-acquired pneumonia (CAP) in healthy, immunized children because of its effectiveness against S. pneumoniae. However, macrolides, which have inferior anti-pneumococcal activity, are the most commonly prescribed class of antibiotics for outpatient CAP. We aimed to determine the comparative effectiveness of β-lactam vs. macrolide antibiotics for the treatment of CAP. METHODS: We conducted a retrospective cohort study in 31 pediatric primary care practices. Patients 3 months to 18 years of age with CAP diagnosed between January 1, 2009 and December 31, 2013 were identified by ICD-9-CM codes. Clinical data were abstracted electronically. Treatment failure was defined as change in antibiotic by the pediatrician, emergency department (ED) visit, or hospitalization for pneumonia in the 2 weeks following diagnosis. Multivariable logistic regression models including children prescribed monotherapy of amoxicillin, broad-spectrum β-lactam antibiotics, or macrolides were built to determine the association of each class with treatment failure, adjusting for clinical and demographic characteristics. RESULTS: Of 10,470 children who received antibiotics for pneumonia, 4252 (40.6%) received amoxicillin, 4459 (42.6%) received macrolides, and 1759 (16.8%) received broad-spectrum β-lactams. The groups differed by age category, proportion of black patients, insurance type, documented fever, ordering of a chest X-ray, and prior antibiotic exposure. Treatment failure occurred in 633 children (6.1%); 418 required a change in antibiotic by the pediatrician, 169 required an ED visit, and 47 required hospitalization. In the adjusted model, macrolide prescribing was associated with a decreased odds of treatment failure in children <5 years old (aOR = 0.52, 95% CI 0.34, 0.78) and in children ≥5 years old (aOR = 0.32, 95% CI 0.25, 0.41). In practices with the lowest macrolide use, this relationship persisted (OR 0.46; 95% CI 0.23, 0.92). CONCLUSION: While rates of treatment failure in children diagnosed with CAP in the outpatient setting were low, macrolides were associated with a lower failure rate than treatment with β-lactams. This may be due to residual confounding by indication or changing epidemiology of outpatient pneumonia. DISCLOSURES: T. Zaoutis, Astellas: Consultant, Consulting fee; Merck: Grant Investigator, Research grant; nabriva: Consultant, Consulting fee.