(1440) Identifying Opportunities to Reduce Fluid Balance in Critically Ill Children

Introduction: Fluid overload (FO), a state of pathologic cumulative fluid balance (CFB), is common in children with critical illness and associated with morbidity and mortality. Key steps to reduce FO are to 1) identify key stakeholders, 2) accurately capture and monitor FB, and 3) identify FO during treatment in a timely manner. While FO is a pervasive problem, different PICU may have unique needs, barriers, and limitations to implementing these steps. We hypothesize that >5% and >10% CFB early in the PICU course are common but practices to address fluid balance vary across sites.

Methods: We utilized a federated data collection framework to facilitate a quality improvement (QI) partnership across 4 PICUs over 6 months. We extracted baseline fluid metric data from the electronic health record (EHR) via analysis scripts developed at a single site and then run locally at each site. We collected prospective data via interviews and surveys with key stakeholders to quantify current practices around FB capture. The primary outcome was %CFB, defined as total intake minus total output divided by PICU admission weight. Chi-square test and Wilcoxon rank sum tests compared results across and within sites.

Results: A total of 3,071 PICU encounters were included. Proportion of encounters with day 1 CFB >5% and >10% did not differ across sites but encounters with day 2 CFB >5% varied from 39 to 54% (p=0.03 across sites) and CFB >10% varied from 16 to 25% (p=0.04 across sites). Count of urine occurrence recordings varied from 16 to 51% and count of patients receiving >100% Holliday Segar fluids administered on Day 1 varied from 20 to 31%; both differed between sites (p < 0.001). Frequency of discussions on rounds about overall FB (42-73%) and specific FB goals (19-39%) varied significantly across sites. However, the most frequent reported barriers to accurate FB capture (involving better recording of oral intake and urine/stool output for ambulatory patients) and achievable opportunities to improve FB capture (e.g. standardizing flush volumes and their recording) were similar across sites.

Conclusions: CFB >5% and >10% was common but variable within the first 2 days across PICUs. Workflow practices to address FO varied significantly across sites despite similar barriers to improve FO and reduce CFB.