144 - Frequency, Characteristics, and Outcomes of Patients Requiring Early PICU Readmission

Poster Number: 144
Publication Number: 144.416

Eleanor A. Sharp, UPMC Childrens Hospital of Pittsburgh, PITTSBURGH, PA, United States; Li Wang, University of Pittsburgh CTSI, Pittsburgh, PA, United States; Jay Berry, Boston Children's Hospital, Boston, MA, United States; Matthew Hall, Children's Hospital Association, Lenexa, Kansas, KS, United States; Catherine Forster, UPMC Childrens Hospital of Pittsburgh, Pittsburgh, PA, United States

Background: Readmission to the Pediatric Intensive Care Unit (PICU) is associated with worse clinical outcomes and higher healthcare costs, but factors associated with PICU readmission within the same hospitalization remain unclear. 

Objective: Describe the prevalence of, and identify factors associated with, early readmission to the PICU during the same hospitalization.

Design/Methods: We performed a retrospective review of clinical and resource utilization data from the Pediatric Health Information System. We included patients age 0-26 years with a PICU admission between January 1, 2016 – December 31, 2019. We defined early readmission as return to the PICU within two calendar days of transfer during the same hospitalization. We excluded patients that died, were discharged, or were transferred to another facility during their index PICU admission in addition to patients that underwent corrective cardiac surgery. We defined medical complexity as the presence of ≥1 complex chronic conditions (CCCs) and technology dependence as presence of ventriculoperitoneal (VP) shunt or tracheostomy. Multivariable modelling was used to analyze associations between patient and clinical variables and our outcome of interest.

Results: 389,219 patient encounters were included. Our population was 55.2% male and had a median age 4 years [IQR: 12-1]. 2.5% of included patients experienced early PICU readmission (Figure 1). On multivariable analysis, factors with highest odds of PICU readmission were presence of any CCCs, with ≥4 CCCs as highest odds of readmission (aOR [95% CI]: 4.17 [3.83–4.53]), parenteral nutrition (2.27 [2.14–2.41]), and presence of VP shunt (1.93 [1.71–2.17]) (Table 1). Factors with decreased odds of PICU readmission included ECMO (0.41 [0.29–0.56]), cardiopulmonary resuscitation (0.79 [0.68-0.92]), and intubation (0.83 [0.79–0.88]). Patients experiencing early PICU readmission had longer overall length of stay (Geometric Mean (Geometric SD): 18.2 (0.9) vs 5.0 (1.1) days, p< .001) and increased odds of mortality (1.7, [1.5-1.9]) (Table 2). Conclusion(s): Although early PICU readmissions are uncommon, they are associated with significantly worse clinical outcomes and increased healthcare utilization.  Patients with medical complexity and technology dependence are especially vulnerable. Identification of variables associated with increased odds of readmission may inform risk-stratification and decision-making when caring for patients recovering from critical illness.
Figure 1. Cohort isolation and categorization of subjects by timing of PICU readmission
Table 1. Multivariable Analysis: Factors Associated with Increased Risk of Early PICU ReadmissionOn multivariate analysis, medical complexity defined by the presence of one or more complex chronic conditions was associated with the highest odds of requiring PICU readmission within two calendar days. The odds of readmission increased sequentially with each additional CCC, with the presence of four or more CCCs conveying the highest risk of readmission seen in our cohort. Technology dependence was also associated with increased risk: patients with ventriculoperitoneal shunts and tracheostomies were 1.93 and 1.38 times as likely to require early PICU readmission, respectively. Regarding characteristics of the index PICU course, patients that received total parenteral nutrition, blood products, sedative medications, and/or nutrition delivery via GJ, GT, or NG tube were at significantly higher risk of early readmission to the intensive care unit.