Discontinuation of Routine Surveillance Cultures During Extracorporeal Membrane Oxygenation in Pediatric Patients: A Single-Center "Before Versus After" Experience, 2022-2025.

OBJECTIVES: To evaluate our "before vs. after" change in practice of stopping routine surveillance cultures in pediatric patients supported with extracorporeal membrane oxygenation (ECMO), by examining patient outcomes, reviewing antimicrobial prescription, and costs. DESIGN: Retrospective before vs. after study. SETTING: PICU, neonatal ICU, and cardiac ICU in a quaternary children's hospital. PATIENTS: Critically ill patients younger than 18 years supported on ECMO between October 2022 and March 2025. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Patients supported on ECMO in the 12 months following the practice change in March 2024 were compared with the same number of ECMO patients from before the practice change (47 in each group). Removal of routine daily blood culture and every-other-day urine and respiratory culture orders in ECMO order sets was associated with a reduction in cultures obtained: from mean (sd) 1.8 (± 0.22) to 0.4 (± 0.19) per ECMO day (p < 0.0001). We failed to identify an associated change in average ECMO run duration (211 vs. 181 hr; p = 0.48) or 30-day mortality (15/47 vs. 15/47). There was an associated decrease in antimicrobial prescriptions, quantified as a percentage of all ECMO days with prescription: (366/414 [88%] vs. 247/356 [69%]; mean difference, 19% [95% CI of the difference 13-25%]; p = 0.002). We estimate that using 2024 prices, there was a cost reduction of $136,000 in the 12 months following the change in practice. CONCLUSIONS: Our experience of introducing in March 2024 a change in using surveillance or scheduled cultures in pediatric ECMO patients in our center is that there was an associated reduction in microbiology cultures, improved antimicrobial stewardship, and cost-savings. In comparison with our experience before the change in practice, we failed to identify any associated negative effects such as increased duration of ECMO support or 30-day survival.