18 - Characteristics and Outcomes of Pediatric Patients Discharged Directly from the Pediatric ICU to Home

Poster Number: 18
Publication Number: 18.201

Katherine S. Schroeder, University of North Carolina at Chapel Hill School of Medicine, Chapel Hill, NC, United States; Margaret Kihlstrom, University of North Carolina at Chapel Hill School of Medicine, Chapel Hill, NC, United States; Ashley G. Sutton, University of North Carolina at Chapel Hill School of Medicine, Chapel Hill, NC, United States; Mercedes A. Bravo, Duke University School of Medicine, Durham, NC, United States; Matthew Pizzuto, University of North Carolina at Chapel Hill School of Medicine, Chapel Hill, NC, United States

Background: Hospital discharge is a complex process requiring coordination of care, follow-up planning and patient/caregiver education. Patients admitted to the pediatric intensive care unit (PICU) are typically transferred to an acute care floor prior to discharge. Various circumstances, including rapid clinical improvement or capacity constraints, may lead to discharge directly from a PICU. This practice has been studied in adult intensive care units and shown to be generally safe, but the incidence of direct PICU discharge for pediatric patients and associated patient characteristics and outcomes are not defined in the literature.

Objective: Our aim was to describe the characteristics and outcomes of patients requiring PICU admission who were discharged directly from the PICU and compare this group to patients who were transferred to a pediatric floor prior to discharge.

Design/Methods: We conducted a retrospective cohort study of patients 0-18 years of age admitted to the PICU at an academic tertiary care center between 1/1/15 and 12/31/20. Patients who were transferred to another facility or died during hospitalization were excluded. Characteristics including age, sex and ethnicity were compared. Need for vasoactive infusion or mechanical ventilation were used as surrogate markers for illness severity. Primary outcomes included length of stay and subsequent hospital encounter within 7 or 30 days.

Results: 4,042 patients were admitted to the PICU during the study period. 768 (19%) were discharged directly from the PICU and 3,274 (81%) were transferred to the floor prior to discharge. Baseline characteristics did not differ between the two groups (Table 1). Patients discharged directly from the PICU were more likely to have required mechanical ventilation (36% vs 25%, p= < 0.01), but less likely to have required a vasoactive infusion (6.5% vs 11%, p= < 0.01) during admission. Discharge directly from the PICU was associated with shorter median length of stay (2.1±3.3 days vs. 5.9±9.4 days, p= < 0.01) without increased frequency of subsequent hospital encounter at 7 or 30 days (p=0.22, p=0.11). Conclusion(s): Discharge of pediatric patients directly from the PICU did not result in increased rate of return to the hospital. Patients discharged from the PICU and those first transferred to the floor were similar demographically, though increased vasopressor requirement and longer length of stay in patients requiring floor transfer may indicate greater degree of illness. Further study is needed to identify optimal patient populations for direct PICU discharge and to understand differences in outcomes for these patients.
Table 1Baseline characteristics of pediatric patients admitted to the PICU and discharged via floor vs. directly from PICU