Empiric Antibiotic Prescription Patterns in Children With Cerebrospinal Fluid Shunt Infection

Infectious Diseases Society of America guidelines recommend vancomycin plus an antipseudomonal beta-lactam (cefepime, ceftazidime or meropenem) in most cases, with adjustments based on local susceptibility patterns. The objective of this study was to determine patterns of inpatient antibiotic use in children undergoing empiric treatment of cerebrospinal (CSF) shunt infection.

Retrospective, observational cohort study at 6 tertiary care children's hospitals among children who developed first shunt infection following initial CSF shunt placement surgery between 2007 and 2012. Primary outcome was use of antimicrobials as empiric therapy, defined as any antibiotic order on the day prior, day of and day after the neurosurgical procedure to treat the infection (ie, occurring before culture results typically guide antibiotic selection). We conducted analyses at the infection and hospital levels from 2007 to 2015.

The 181 children who received empiric antibiotics most received vancomycin (n = 155, 85.6%), cefazolin (n = 65, 35.9%), ceftriaxone (n = 55, 30.4%), cefotaxime (n = 43, 23.8%) and cefepime (n = 36, 19.9%). Few children received combinations of empiric antibiotic combinations recommended by the Infectious Diseases Society of America guidelines: vancomycin and cefepime (n = 7, 3.6%), vancomycin and ceftazidime (n = 0, 0%) or vancomycin and meropenem (n = 1, 0.5%). There was no evidence indicating a change in antimicrobial coverage over the study period.

Empiric therapy for CSF shunt infections represents a key opportunity for antimicrobial stewardship programs to ensure guideline concordance and promote appropriate antibiotic use.