LAUSR

Introduction/Hypothesis: Fluid overload is associated with worse clinical outcomes in critically ill children. We aimed to describe the absolute and relative volumes of each fluid source accumulated during the first 72 hours of critical illness and study the association of fluid overload with outcomes. Methods: We performed a retrospective cohort analysis of all pediatric intensive care unit (PICU) patients admitted to a large, urban, academic PICU between 9/2010 and 8/2018 who were alive and in the hospital 72 hours after admission. Total fluid input and source of fluid were calculated for the first 72 hours of admission. Fluids were categorized into: resuscitation, maintenance, blood products, nutrition, and fluid creep (i.e. electrolytes, medications, and small volumes for keeping lines open). Fluid overload (FO) was estimated using fluid balance and weight. We analyzed the inhospital mortality and ventilatorfree days at 28 days associated with FO>15% after adjusting for age and PRISM III score using logistic and Poisson regression, respectively. Results: 11,176 patient encounters met inclusion criteria with a mortality rate of 2.1%. 1,019 (9.1%) patients had FO>15% at 72h, and had a mortality of 5.2%. The mean FO at 3 days for all patients was 5.7% ± 9.2%. The mean fluid input in the first 72h for all patients was 239 ml/kg (SD ± 137 ml/kg) and was composed of 49% maintenance, 42.6% nutrition, 4.6% fluid creep, 3.3% resuscitation, and 1.1% blood products. The mean FO at 3 days for patients with FO>15% was 22.4% ± 17%. The mean fluid input in the first 72h for patients with FO>15% was 429 ml/kg (SD ± 213 ml/kg) and was composed of 44% maintenance, 42.6% nutrition, 6.5% fluid creep, 5.6% resuscitation, and 1% blood products. After adjusting for severity of illness and age, FO>15% at 72h was associated with an adjusted odds ratio = 2 (95% confidence interval [CI] 1.4, 2.8) for mortality and an adjusted incidence rate ratio = 0.86 (p<0.001) for ventilator-free days. Conclusions: Fluid overload at 72h is independently associated with higher mortality and lower likelihood of having a ventilator-free day. 44% of fluid intake in patients with fluid overload in the first 72 hours is maintenance and only 5.6% is resuscitation. Fluid restrictive strategies focusing on reducing maintenance fluids warrants further study.