O0436 Oct.5: 3:15pm - Fluid Creep in the PICU: Characterizing Fluid Administration Beyond Maintenance Fluids

Critically ill children receive fluids for multiple reasons including resuscitation, nutrition, and medications. While it is well-established that fluid overload (FO) is associated with negative outcomes in critically ill children, it is challenging for clinicians to consider all fluid sources and adjust accordingly. “Fluid creep” (FC) refers to insidious IV fluid intake (medications, flushes, etc.) in excess of standard predicted maintenance fluid requirements (MFR), and can represent a major source of fluid intake. This phenomenon has previously been described in burn patients and adults, but is less well-established among pediatric patients. The prevalence of FC and whether FC contributes to fluid overload in critically ill children is unknown.

Analysis included EHR-derived data from all PICU patients admitted between 2010-2019 and was limited to the first 120 hours of PICU stay. Fluid volumes were indexed to patient weight and categorized by type as either IV fluids and medications, enteral intake, blood products, or other. IV fluids were deemed “resuscitation” if administered at ≥10mL/kg/hr. The Holliday-Segar method was used to estimate MFR. FC was defined as non-resuscitation IV fluid in excess of MFR. FO was defined as ≥100ml/kg(10%) in cumulative fluid balance at any time. A multivariable logistic regression model was used to quantify the association between Day 1 fluid creep and fluid overload. Because sicker patients could be expected to receive additional fluids and medications, initial severity of illness as measured by the Day 1 Pediatric Logistic Organ Dysfunction Score-2 (PELOD-2) was added as a covariable.

The study included 14176 admissions and 33077 ICU days, involving 10148 distinct patients. The data showed average daily fluid intake was 97.9mL/kg(SD=76) of which 60.1mL/Kg(61%) was IV fluids and medications, 32.6mL/kg(33%) was enteral, 4.2mL/kg(4.3%) was blood products, and 1.1mL/kg(1.1%) was other. Of the IV fluids and medications, 6.5mL/kg(6.6%) was represented by resuscitation. Average FC on Day 1 was 23.4mL/kg(SD=31), and FC represented 21% of all fluid intake through Day 5. Average fluid intake for each ICU day, stratified by type, is shown in Figure 1. FO was present at some point in 16% of admissions. The logistic regression model showed that day 1 FC was associated increased odds of FO (OR 1.23, 95%CI 1.21-1.24, p<0.001) independent of PELOD-2 score (Table 1).

The results suggest that fluid creep is a significant component of overall fluid intake in the PICU, and increased FC early in the PICU stay is associated with fluid overload independent of major organ dysfunction. Early and ongoing consideration of all sources of fluid administration is required to avoid excessive fluid intake and associated fluid overload. Given the proven association between fluid overload and increased mortality, future studies should evaluate the potential impact of avoiding fluid creep on outcomes in the PICU.