Antibiotics for Aspiration Pneumonia in Neurologically Impaired Children

Abstract

Objective: To compare hospital outcomes associated with commonly used antibiotic therapies for aspiration pneumonia in children with neurologic impairment (NI).

Design/methods: A retrospective study of children with NI hospitalized with aspiration pneumonia at 39 children's hospitals in the Pediatric Health Information System database. Exposure was empiric antibiotic therapy classified by antimicrobial activity. Outcomes included acute respiratory failure, intensive care unit (ICU) transfer, and hospital length of stay (LOS). Multivariable regression evaluated associations between exposure and outcomes and adjusted for confounders, including medical complexity and acute illness severity.

Results: In the adjusted analysis, children receiving Gram-negative coverage alone had two-fold greater odds of respiratory failure (odds ratio [OR] 2.15; 95% CI: 1.41-3.27), greater odds of ICU transfer (OR 1.80; 95% CI: 1.03-3.14), and longer LOS [adjusted rate ratio (RR) 1.28; 95% CI: 1.16-1.41] than those receiving anaerobic coverage alone. Children receiving anaerobic and Gram-negative coverage had higher odds of respiratory failure (OR 1.65; 95% CI: 1.19-2.28) than those receiving anaerobic coverage alone, but ICU transfer (OR 1.15; 95% CI: 0.73-1.80) and length of stay (RR 1.07; 95% CI: 0.98-1.16) did not statistically differ. For children receiving anaerobic, Gram-negative, and P. aeruginosa coverage, LOS was shorter (RR 0.83; 95% CI: 0.76-0.90) than those receiving anaerobic coverage alone; odds of respiratory failure and ICU transfer rates did not significantly differ.

Conclusions: Anaerobic therapy appears to be important in the treatment of aspiration pneumonia in children with NI. While Gram-negative coverage alone was associated with worse outcomes, its addition to anaerobic therapy may not yield improved outcomes.

FIG 1

Variation in the Spectrum of Empiric Antimicrobial Coverage across Hospitals. Each column represents data from one hospital. The spectrum of empiric antimicrobial coverage varied vastly across hospitals: anaerobic coverage was prescribed for 0%-44% of patients; Gram-negative coverage alone was prescribed for 3%-26% of patients; anaerobic and Gram-negative coverage was prescribed for 25%-90% of patients; and anaerobic, Gram-negative, and P. aeruginosa coverage was prescribed for 0%-65% of patients. ^aGram-negative denotes coverage against Gram-negative pathogens except for P. aeruginosa

FIG 2

Adjusted outcomes. Independent effect of antimicrobial spectra of activity on outcomes (adjusted odds ratio and 95% CI for acute respiratory failure and ICU transfer; adjusted rate ratio for the length of stay) for children with NI hospitalized with aspiration pneumonia. Anaerobic therapy alone served as the referent group. Generalized linear mixed-effects models included hospital as a random effect to account for clustering by hospital and individual covariates (ie, age, presence of complex chronic condition diagnoses (in neurologic/neuromuscular, gastrointestinal, congenital/genetic defect, respiratory, cardiovascular, metabolic, neonatal, renal, malignancy, and hematology/immunology categories), the number of complex chronic conditions, technology dependence (any and gastrointestinal), performance of diagnostic tests on presentation (blood culture, respiratory culture, urine culture, and chest radiograph), and markers of severe illness on presentation (pleural drainage, receipt of blood products, and receipt of vasoactive agents)). ICU admission was included as a covariate in acute respiratory failure and length of stay outcome models. Abbreviations: ICU, intensive care unit; NI, neurologic impairment.