533 - Association of Hypothermia With Mortality Among Children in the Emergency Department: A Retrospective Cross-Sectional Study

Publication Number: 2518.533

Lucy Campbell, Northwestern Feinberg School of Medicine, Chicago, IL, United States; Aarthi Kottapalli, Northwestern University The Feinberg School of Medicine, Chicago, IL, United States; Leslie A. Hueschen, University of Missouri-Kansas City School of Medicine, Kansas City, MO, United States; Jonathan R. Higgins, University of Colorado Anschutz School of Medicine, Aurora, CO, United States; Erin J. Meyer, UMass Memorial Children's Medical Center, Natick, MA, United States; Laura Sartori, Childrens Hospital of Philadelphia, Philadelphia, PA, United States; Paria M. Wilson, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States; Naghma S. Khan, Children’s Healthcare of Atlanta, Atlanta, GA, United States; Alexander Rogers, University of Michigan Medical School, Ann Arbor, MI, United States; Sriram Ramgopal, Ann & Robert H. Lurie Children's Hospital of Chicago, CHICAGO, IL, United States

Background: Hypothermia is a risk factor for poor outcomes in infants and children with sepsis or trauma. However, the overall association of hypothermia and mortality among all children in the emergency department (ED) is unknown.

Objective: To evaluate whether hypothermia is an independent risk factor for in-hospital mortality among children presenting to the ED.

Design/Methods: We performed a retrospective multicenter cohort study of pediatric ( < 18 years) ED encounters from 2012 to 2021 using the Pediatric Emergency Care Applied Research Network public use registry. We evaluated for the association between first-recorded temperature (as hypothermia [≤36°C], euthermia, or fever [≥38°C]) and in-hospital mortality in multivariable logistic regression models adjusted for sex, age, vital signs (respiratory rate, heart rate, and systolic blood pressure), Emergency Severity Index, Child Opportunity Index, season, diagnosis, and medical complexity. We evaluated the association between temperature and mortality within subgroups defined by the presence of trauma, age < 90 days, medical complexity, and exclusion of in-ED mortality. We evaluated the continuous association of initial temperature with mortality using a restricted cubic spline function.

Results: We included 5,308,892 encounters (median age 4.92 years, IQR 1.68-10.7; 52.5% male). Hypothermia and fever were present in 161,111 (3.0%) and 715,366 (13.5%), respectively (Table 1). In-hospital mortality occurred in 1,918 (0.04%) encounters; 39.2% hypothermic, 47.4% euthermic, and 13.3% febrile. In the multivariable model, triage hypothermia was associated with mortality (adjusted odds ratio [aOR] 8.67, 95% confidence interval [CI] 7.66-9.80; Table 2). Fever was not associated with mortality, except among infants < 90 days (aOR 2.41, 95% CI 1.36-4.29). Hypothermia was associated with mortality in all subgroup analyses, including trauma (aOR 18.85, 95% CI 11.51-30.87), young infants (aOR 14.19, 95% CI 9.92-20.29), presence (aOR 4.77, 95% CI 4.13-5.51) and absence (aOR 20.88, 95% CI 15.50-28.14) of medical complexity, and exclusion of patients with in-ED mortality (aOR 6.28, 95% CI 5.46-7.23). Visualized as a restricted cubic spline, there was a higher probability of in-hospital mortality with lower initial temperatures, with an optimal cutpoint for this outcome identified at 35.9°C (Figure 1).

Conclusion(s): Hypothermia is an independent risk factor for mortality among children in the ED, overall and within subgroups. Hypothermia may be an important measure to identify and triage children at highest risk of poor outcomes.

Table 1: Patient characteristics
Footnote: Numbers represent N (%) or median [IQR]

Table 2: Multivariable associations of hypothermia and fever with in-hospital mortality, with euthermic patients as reference
Footnote: Models adjusted for all the following factors when applicable: sex, age (except for the age <90 days subgroup), respiratory rate, systolic blood pressure, heart rate, ESI triage category, COI, season, DGS category (except for the trauma and non-trauma diagnosis subgroups), type of complex chronic condition (except for the absence of medical complexity subgroup)

Figure 1: Continuous association of initial presenting temperature with in-hospital mortality