.jpg "Jaskaran Rakkar, MD photo")
Jaskaran Rakkar, MD
UPMC Children's Hospital of Pittsburgh
Pittsburgh, Pennsylvania
Disclosure information not submitted.
.jpeg.jpg "Jonathan Pelletier, MD photo")
Jonathan Pelletier, MD
Clinical Instructor
Children's Hospital of Pittsburgh of University of Pittsburgh Medical Center
Pittsburgh, Pennsylvania
Disclosure information not submitted.
Dennis Simon, MD
Assistant Professor of Critical Care Medicine and Pediatrics
UPMC Children's Hospital of Pittsburgh, United States
Disclosure information not submitted.
Alicia Au, MD
Assistant Professor, Critical Care Medicine. Associate Medical Director, PICU
Children's Hospital of Pittsburgh of UPMC, United States
Disclosure information not submitted.
.jpg "Patrick Kochanek, MD, MCCM photo")
Patrick Kochanek, MD, MCCM
Director, Safar Center for Resuscitation Research
UPMC Presbyterian
Pittsburgh, Pennsylvania
Disclosure information not submitted.
Gilles Clermont, MD, MS
Professor CCM, Mathematics, Clinical Translational Sci and IE
VA Pittsburgh Health System
Pittsburgh, Pennsylvania
Disclosure information not submitted.
Robert Clark, MD, FCCM
Professor of Critical Care Medicine and Pediatrics
Childrens Hospital of Pittsburgh of UPMC, United States
Disclosure information not submitted.
Christopher Horvat, MD, MHA
Children's Hospital of Pittsburgh of UPMC
Pittsburgh, Pennsylvania
Disclosure information not submitted.
Title: Identifying Causal Relationships between ICP and Intensity of Medical Therapies in Pediatric TBI
Introduction: Severe pediatric traumatic brain injury (TBI) is a dynamic disease process and intracranial pressure (ICP)-directed management is currently the principal component of post-TBI critical care. We aimed to determine the relationships between ICP, ICP-directed medical therapeutic support, and in-hospital mortality.
Methods: Ten years of data (1/2009-4/2019) were extracted from the electronic medical record (EMR) for patients ≤18 years old with severe TBI admitted to our pediatric intensive care unit. We developed an EMR-based Pediatric Intensity Level of Therapy scale (ePILOT) model to quantify ICP-directed therapy that included CSF diversion, hyperosmolar therapies, sedation/analgesia, barbiturates, PaCO2, neuromuscular blockade, hypothermia, and serum Na+levels. Using a stepwise approach, we constructed a Mixed Graphical Model and then a multivariable logistic regression to determine scores for these therapies. Data were summarized with descriptive statistics and temporal analyses performed for the first 7 days of hospitalization. Directed Acyclic Graph (DAG) analysis was used to determine causal relationships.
Results: 135 ICP-monitored TBI patients (6.9±5.6y; 65% male; admission Glasgow Coma Scale [GCS] score 4 [3-7]; mortality 16%, hospital length of stay 20.7±14.0 d) were analyzed. The mixed graphical model identified associations amongst different variables, as well as between variables and ICP. The median ePILOT score and mean ICP were lower in survivors vs. non-survivors (p < 0.05). Time-series analyses showed that patients who died had a higher ePILOT score in the first 7 days of hospitalization. DAG analysis revealed a directed path between ICP and ePILOT, i.e. changes in ICP caused changes in ICP-directed medical therapies, which then influenced mortality. A similar causal relationship was also identified between admission GCS and ePILOT score.
Conclusions: In this pediatric TBI patient dataset, the ePILOT model captures the intensity of ICP-directed medical therapies. There is a causal relationship between ICP and ePILOT score impacting mortality, suggesting a reactionary response to ICP in terms of ICP-directed therapies at our center. Further study to determine the generalizability of the ePILOT model and to identify a means for proactive vs. reactive management of ICP appears warranted.