To evaluate a guideline for antibiotic decisions in children with suspected ventilator-associated infection.
Prospective, observational cohort study conducted in 22 PICUs in the United States and Canada.
PICUs in 22 hospitals from April 2017 to January 2019.
Children less than 3 years old on mechanical ventilation greater than 48 hours who had respiratory secretions cultured and antibiotics initiated for suspected ventilator-associated infection.
After baseline data collection in children with suspected ventilator-associated infection (Phase 1), a consensus guideline was developed for advising antibiotic continuation or stopping at 48–72 hours (Phase 2) and implemented (Phase 3). Guideline-based antibiotic recommendations were provided to the treating clinicians once clinical and microbiologic data were available. Demographic and outcome data were collected, and guideline compliance and antibiotic utilization evaluated for Phase 1 and Phase 3.
MEASUREMENTS AND MAIN RESULTS:
Despite education and implementation efforts, guideline-concordant antibiotic management occurred in 158 of 227 (70%) Phase 3 subjects compared with 213 of 281 (76%) in Phase 1. Illness severity and positive respiratory cultures were the primary determinants of antibiotic continuation. For subjects with a positive respiratory culture but a score for which antibiotic discontinuation was recommended (score ≤ 2), only 27% of Phase 3 subjects had antibiotics discontinued. Antibiotic continuation was not associated with improved outcomes in these subjects and was associated with significantly longer duration of ventilation (median 5.5 d longer) and PICU stay (5 d longer) in the overall study population. Positive respiratory cultures were not associated with outcomes irrespective of antibiotic treatment.
Antibiotic guideline efficacy and safety remain uncertain due to clinician failure to follow the guideline, instead primarily relying on respiratory culture results. Strategies to overcome clinician perceptions of respiratory cultures and other barriers will be vital for improving guideline adherence and antibiotic use in suspected ventilator-associated infection in future studies.