Marc Dadios, MD
Harbor UCLA Medical Center
Torrance, California
Disclosure information not submitted.
Daniel Kim, MD
Pediatric Residency Program
University of California, Irvine, United States
Disclosure information not submitted.
Christopher Babbitt, MD
Pediatric Intensivist
MemorialCare Miller Childrens Hospital and Womens Hospital Long Beach
Long Beach, California
Disclosure information not submitted.
Jeffrey Johnson, MD, MA
Medical Director of Pediatric Inpatient and Critical Care Services
LAC & USC Medical Center, United States
Disclosure information not submitted.
Brant Putnam, MD, FACS
Chief, Division of Trauma & Acute Care Surgery / Surgical Critical Care
Harbor-UCLA Medical Center
Torrance, United States
Disclosure information not submitted.
Robert Hateley, MD
Pediatric Residency Program
Harbor-UCLA Medical Center, United States
Disclosure information not submitted.
Dennys Estevez
Senior Statistician
Lundquist Institute, United States
Disclosure information not submitted.
Richard Mink, MD, MACM,FCCM
Director, Pediatric Critical Care Fellowship
Harbor UCLA Medical Center, United States
Disclosure information not submitted.
Title: Do All Children With Traumatic Intracranial Hemorrhage Need to be Monitored in the Pediatric ICU?
Introduction: Children with traumatic intracranial hemorrhage (tICH) on initial head CT are usually admitted to the pediatric intensive care unit (PICU) for neuromonitoring but this may not be necessary. Three recent studies have proposed decision trees for patients with tICH to identify those who can be admitted to non-PICU locations but validation is needed before they can be utilized. This study tested the ability of these algorithms to correctly identify pediatric patients with tICH who are at low risk for requiring a critical care intervention (CCI) after admission.
Methods: This retrospective study included all patients < 18 years admitted from 3 level 1 trauma centers between Jan 1, 2010 and Sept 1, 2019 who had tICH on initial head CT and did not require a CCI prehospital or in the emergency department. We applied our data to the Children’s Intracranial Injury Decision Aid (CHIIDA) scoring system (Greenberg et al.) and to the decision trees proposed by Ament et al. and Burns et al. A CCI was defined as requirement for assisted ventilation, intubation, hyperosmolar therapy, vasoactive medications, cardiopulmonary resuscitation, arrhythmia management, massive blood transfusion, neurosurgical intervention, and/or placement of an intracranial pressure monitor. We assigned CHIIDA scores and tested the two clinical decision trees using our data.
Results: 2819 patients with traumatic brain injury were identified within this period. 646 patients with tICH who did not require a preadmission CCI were included in the study and of these, 50 (7.7%) received an inpatient CCI. Patients with a CHIIDA score of 0 had a 2.0% risk of requiring an inpatient CCI compared with 2.6% in Greenberg’s study. Using the decision tree by Ament, 4.2% were classified as low risk versus 1.9% in their study while the decision tree by Burns identified 2.0% as low risk compared with 0.6% in their study. Based on the algorithms of Ament and Burns, 18 and 10 patients would have been characterized as low risk, respectively, but actually had an inpatient CCI.
Conclusions: This study suggests that the CHIIDA scoring system can identify tICH patients at low risk for requiring an inpatient CCI. Those with a CHIIDA score of 0 may be monitored out of the PICU. This study highlights the importance of validating proposed algorithms before utilization.