O1018: Decreasing Antimicrobial Use in Febrile Young Infants

Purpose/Objectives: In 2021, the American Academy of Pediatrics (AAP) published a clinical practice guideline (CPG) recommending using inflammatory markers (IMs) including procalcitonin (PCT), absolute neutrophil count (ANC), and C-reactive protein (CRP) to identify febrile infants at risk for invasive bacterial infections (IBI) in need of antimicrobials. Our aim was to decrease the percentage of febrile infants 22-56 days old with normal IMs and urinalysis (UA) receiving antimicrobials from a baseline of 23% to 10% within one year of a quality improvement initiative.

Design/Methods: A multidisciplinary team reviewed the AAP CPG, local epidemiology, and identified key drivers including education, pathway updates, and clinical decision support (CDS).

At baseline, infants 22-28 days old were admitted on antibiotics. Infants 29-56 days old were admitted on antibiotics if they had an abnormal UA, white blood cell count (WBC), or band to neutrophil ratio. In the intervention period, infants 22-56 days old were admitted on antibiotics if they had abnormal IMs or UA.

Infants 22-56 days old from January 2017 – February 2022 (baseline) and March 2022 – February 2023 (intervention) with a temperature >100.4°F at home or in the ED and blood, urine, or cerebrospinal fluid cultures obtained in the ED were included.

Measures:
Our outcome measure was rates of antimicrobial administration in infants with normal blood biomarkers and UA based on institutional standards at the time. Process measures included percentage of infants with resulted WBC and PCT. Balancing measures were missed IBI (patients with IBI not empirically treated in the ED), 72-hour revisits, and ED length of stay (LOS). Statistical process control methodology was used to assess these interventions.

Interventions:
PDSA Cycle 1: Education, updated clinical pathway, developed age-based non-interruptive CDS. Abnormal IMs defined as: PCT >0.5 ng/mL, CRP >2 mg/dL, or ANC < 1000 or >4000 neutrophils/µL. Abnormal UA defined as positive gram stain or >10 WBC/high power field (HPF).

PDSA Cycle 2: Updated definition of positive UA to any leukocyte esterase or >5 WBC/HPF.

PDSA Cycle 3: Removed CRP as an IM.

Results: Rates of antimicrobial administration in infants with normal blood biomarkers and UA decreased from 23% to 5% with a downward centerline shift (Figure 1). Percentage of infants with resulted ANC maintained at 97% and PCT increased from 5% to 97%, causing an upward centerline shift (Figure 2). Pre-intervention, there were 3 patients with missed bacteremia but none after our intervention. In both time periods, there were no missed cases of bacterial meningitis. There was no significant change in 72-hour revisits or ED LOS.

Conclusion/Discussion: This quality improvement project resulted in a rapid increase in the utilization of IMs and safely decreased the use of antimicrobials in febrile young infants without missing IBIs, increasing 72-hour revisits, or prolonging LOS.