Nadir Yehya, MD, MSCE
Children's Hospital of Philadelphia
Cherry Hill, NJ
Disclosure information not submitted.
Julie Fitzgerald, MD, PhD, FCCM
Children's Hospital of Philadelphia
Philadelphia, Pennsylvania
Disclosure information not submitted.
Katie Hayes, BS
Lead Clinical Data Management Specialist in the Pediatric Sepsis Program
Children's Hospital of Philadelphia, United States
Disclosure information not submitted.
Donglan Zhang, BS
Scientist
The Children's Hospital of Philadelphia, United States
Disclosure information not submitted.
Jenny Bush, RN, BSN
Research Coordinator
The Children's Hospital of Pennsylvania, United States
Disclosure information not submitted.
Natalka Koterba
Research Specialist
University of Pennsylvania, United States
Disclosure information not submitted.
Fang Chen, PhD
Scientist
Center for Cellular Immunotherapies, Perelman School of Medicine, Univers, United States
Disclosure information not submitted.
Florin Tuluc, MD, PhD
Scientist
Children's Hospital of Philadelphia
Philadelphia, Pennsylvania, United States
Disclosure information not submitted.
David Teachey, MD
Professor
Childrens Hospital of Philadelphia, United States
Disclosure information not submitted.
Fran Balamuth, MD, PhD, MSCE, MD, PhD, MSCE
Associate Director of Research in the Emergency Department
Children's Hospital of Philadelphia, United States
Disclosure information not submitted.
Simon Lacey, PhD
Professor
Center for Cellular Immunotherapies, Perelman School of Medicine, University of, United States
Disclosure information not submitted.
Joseph Melenhorst, PhD
Professor
University of Pennselvania, United States
Disclosure information not submitted.
Scott Weiss, MD, FCCM
Associate Professor of Anesthesiology, Critical Care, and Pediatrics
The Children's Hospital of Philadelphia
Glen Mills, Pennsylvania, United States
Disclosure information not submitted.
Title: Temperature Trajectory Sub-phenotypes and the Immuno-Inflammatory Response in Pediatric Sepsis
Introduction: Enrichment strategies for clinical trials using biomarker-based subtyping has been difficult to operationalize in sepsis. As an alternative approach, four sub-phenotypes defined by distinct temperature trajectories have been reported in adult sepsis. Given the distinct epidemiology of pediatric sepsis, we aimed to classify septic children into de novo sub-phenotypes derived from temperature trajectories and compare cytokine, immune function, and immunometabolic markers across subgroups. We hypothesized that sub-phenotypes based on temperature trajectories would be associated with divergent biology.
Methods: This was a secondary analysis of a prospective cohort of 191 septic children from a single academic pediatric intensive care unit. We performed group-based trajectory modeling (GBTM) using available temperatures over the first 72 hours of sepsis to identify latent de novo profiles. We then used mixed effects regression to determine if the resulting temperature trajectory-defined sub-phenotypes were associated with cytokines, immune function (absolute lymphocyte count, ex vivo LPS-stimulated TNFα, and mHLA-DR expression), and mitochondrial respiration from measurements over the first 14 days after sepsis onset.
Results: There were 7,866 temperature measurements available, with a median (interquartile range) of 28 (IQR 23, 57) measurements per patient. We identified four sub-phenotypes: hypothermic, normothermic, hyperthermic fast-resolvers, and hyperthermic slow-resolvers. Hypothermic patients were less often previously healthy and exhibited lower levels of pro- (TNFα, IL-1β, IL-6, IL-12, G-CSF, and sIL2R) and anti-inflammatory (IL-10 and IL1Ra) cytokines. There were no differences in the immune function or peripheral blood mitochondrial respiration. Hospital mortality was higher in hypothermic children (17% versus 3 to 11%), but this did not reach statistical significance.
Conclusions: Critically ill septic children can be categorized into temperature trajectory-defined sub-phenotypes that parallel adult sepsis. Hypothermic children exhibit a blunted cytokine profile. GBTM has utility for identifying subtypes of clinical syndromes by incorporating readily available longitudinal data, rather than relying on inputs from a single timepoint.