485 - Blood culture indications in critically ill children: multi-center exploration of current clinical practices to optimize a diagnostic stewardship program

Publication Number: 485.3705

Charlotte Woods-Hill, Perelman School of Medicine at the University of Pennsylvania, Wilmington, DE, United States; Aaron Milstone, Johns Hopkins University School of Medicine, Baltimore, MD, United States; Ryan H. Burnett, Childrens Hospital of Philadelphia, New York, NY, United States; Chris Bonafide, Childrens Hospital of Philadelphia, Philadelphia, PA, United States; Margaret Robinson, University of California, San Francisco, School of Medicine, San Francisco, CA, United States; Natalie Z. Cvijanovich, University of California, San Francisco, School of Medicine, Oakland, CA, United States; Su Jin Joo, UCSF Benioff Children's Hospital Oakland, San Francisco, CA, United States; Anita I. Sen, Columbia University Vagelos College of Physicians and Surgeons, New York, NY, United States; Irene Frantzis, Columbia University Vagelos College of Physicians and Surgeons, New York, NY, United States; Karen Ravin, NemoursAlfred I. duPont Hospital for Children, Wilmington, DE, United States; Kimberly McMahon, Nemours Children's Hospital, Wilmington, DE, United States; Ishminder Kaur, University of California, Los Angeles David Geffen School of Medicine, Los Angeles, CA, United States; Myke Federman, UCLA Mattel Childrens Hospital, Los Angeles, CA, United States; Courtney Frye, Riley Hospital for Children at IU School of Medicine, Division of Critical Care, INDIANAPOLIS, IN, United States; Jack G. Schneider, Indiana University School of M, Indianapolis, IN, United States; Brittany Player, Medical College of Wisconsin, Wauwatosa, WI, United States; Amy Goza, Children's Hospital of Wisconsin, Milwaukee, WI, United States; Kari Rajzer- Wakeham, Children's Hospital of Wisconsin, Waukesha, WI, United States; Kris A. Bryant, University of Louisville, Louisville, KY, United States; Melissa B.. Porter, Norton Children's, Louisville, KY, United States; Meghan Gray, NewYork-Presbyterian Morgan Stanley Children's Hospital, New York, NY, United States; Lynn Ramirez-Avila, UCSF Benioff Children's Hospital San Francisco, San Francisco, CA, United States

Background: Overuse of blood culture testing can lead to false positive results and related harms, including unnecessary antibiotic administration. Blood culture practices in the pediatric intensive care unit (PICU) vary widely. Understanding current practice is critical prior to setting targets for safe and feasible blood culture reduction through diagnostic stewardship efforts.

Objective: Our objective was to characterize blood culture practice patterns in multiple PICUs prior to implementation of a diagnostic stewardship program designed to optimize culture use.

Design/Methods: We enrolled 8 PICUs in the Bright STAR: LIBRA trial to study the impact of different strategies to implement standardized blood culture practices and reduce culture overuse. Each site was asked to audit 120 randomly selected blood cultures from PICU patients during a 12-month pre-intervention baseline period, and collect data on culture indications, sources, and patient characteristics.

Results: 931 blood cultures were audited across 8 sites (median 116 cultures/site, range 108-121). Positive blood cultures represented 6% of the sample. The majority of cultures (70%) were for new symptoms, such as fever not present in prior 48 hours; while 23% were for persistent symptoms, and 7% were screening cultures in asymptomatic patients. Of the 64 cultures taken in asymptomatic patients, two were positive (3%). There was a documented concern for sepsis as the trigger for the blood culture in a minority (37%) of patients with new symptoms, and a minority (25%) of patients with persistent symptoms. 100% of blood cultures drawn for new symptoms that ultimately returned positive for bacteria had a documented concern for sepsis at the time of culture. Over half of audited cultures (51%) were obtained when another source of fever was suspected or documented. Culturing multiple lumens of a central venous line (CVL), if one was present, was relatively uncommon but did occur (12% of 137 CVL cultures). For repeat blood cultures, 39% were taken at 24 hours and 61% were taken at 48 hours.

Conclusion(s): A large proportion of blood cultures in PICU patients were drawn for new fever, though often without a clear concern for sepsis and often when another source of fever was suspected or documented. Clinician concern for sepsis was uniformly documented for positive cultures sent for new symptoms. Practices were variable regarding timing of repeat cultures and sources. Work to develop and optimize thresholds and standard practices for sending cultures is critical to inform ongoing diagnostic stewardship efforts.