1339 - Antibiotic Prescribing Varies among Patients with a Documented Penicillin or Cephalosporin Adverse Drug Reaction

Background: Studies have shown that over half of hospitalized children receive an antibiotic during their encounter, of which between 30-50% is considered inappropriate. Antibiotic prescribing is further complicated as approximately 10% of children are labeled beta-lactam allergic, resulting in the use of either broad-spectrum or suboptimal therapy.  The purpose of this study was to compare antibiotic prescribing between patients with a documented ADR vs. those without using a nationwide sample of hospitalized children.

Methods: We performed a point prevalence study among 32 hospitals between July 2016-December 2017 where data were collected via chart review on pediatric patient and antimicrobial characteristics, including the indication for all antimicrobials. In additional, ADR history data were collected on which antimicrobial(s) were documented (e.g., penicillin, cephalosporins).  Patients were mutually assigned into either: 1) no documented ADR; 2) penicillin ADR-only; 3) cephalosporin ADR-only; and 4) ADR for both penicillin and cephalosporin.  The distribution of antibiotics were compared between the ADR groups, stratified by the indication for treatment.

Results: A total of 12,250 pediatric patients (17,929 antibiotic orders) who were actively receiving antibiotics were identified.  A history of penicillin and cephalosporin ADR was documented in 5.5% and 2.8% of these patients, respectively.  When compared to patients with no documented ADR, penicillin ADR patients were more likely to receive a fluoroquinolone for a SSTI infection (odds ratio [OR]: 5.6), surgical prophylaxis (OR: 18.8) or for surgical treatment (OR: 5.2) (see Figure). Conversely, penicillin ADR patients were less likely to receive first-line agents, such as narrow-spectrum penicillin for bacterial LRTI (OR: 0.08) and piperacillin/tazobactam for GI infections (OR: 0.22). Cephalosporin ADR patients exhibited similar patterns with increased use of carbapenems and fluoroquinolones when compared to patients with no ADR.Figure 1: Odds of Receiving Select Antimicrobials Among PCN ADR Patients When Compared to Non-ADR patients, by Indication


Conclusion: A large, nationwide sample of pediatric patients who were actively prescribed antibiotics helped identify several diagnoses where comprehensive guidelines for appropriate ADR prescribing and increased ADR de-labeling initiatives are needed to ensure optimal treatment.