60 - Validation of the PECARN Traumatic Brain Injury Clinical Prediction Rule

Poster Number: 60
Publication Number: 60.104

Nathan Kuppermann, University of California, Davis, School of Medicine, Sacramento, CA, United States; Mohamed Badawy, University of Texas Southwestern Medical School, Dallas, TX, United States; Pradip P. Chaudhari, Children's Hospital Los Angeles, Los Angeles, CA, United States; Paul Ishimine, University of California, San Diego School of Medicine, San Diego, CA, United States; Irma Ugalde, McGovern Medical School at the University of Texas Health Science Center at Houston, Houston, TX, United States; Kevan A. McCarten-Gibbs, University of California, San Francisco, School of Medicine, Oakland, CA, United States; Kenneth Yen, University of Texas Southwestern Medical School, Dallas, TX, United States; Nisa S. Atigapramoj, University of California, San Francisco, School of Medicine, San Francisco, CA, United States; Donovan Nielsen, University of California, Davis, School of Medicine, Sacramento, CA, United States; Daniel J. Tancredi, University of California, Davis, School of Medicine, Sacramento, CA, United States; P. David Adelson, Phoenix Children's Hospital, Phoenix, AZ, United States; James F. Holmes, University of California, Davis, School of Medicine, sacramento, CA, United States

Background:

Background: The two Pediatric Emergency Care Applied Research Network (PECARN) traumatic brain injury (TBI) prediction rules were derived to assist clinicians with decision-making regarding computed tomography (CT). Each consists of six factors, and if none are present, the child is very low risk for clinically important TBI (ciTBI).

Objective:

Objective: The objective of this study was to externally validate these prediction rules for identifying children at very low risk of ciTBI for whom CT could be avoided.

Design/Methods: Methods: We prospectively enrolled children ( < 18 years) with minor blunt head trauma (Glasgow Coma Scale scores 14-15), at six emergency departments (EDs) to validate the accuracy of the two age-based (0- < 2 years and 2-18 years) TBI prediction rules in a new multicenter cohort. Clinicians documented if the prediction rule was positive or negative prior to CT, if obtained. ciTBI was defined as a TBI with any of the following: death from the TBI, neurosurgery, intubation for ≥ 24 hours, or hospitalization for ≥2 nights for the TBI in association with a positive CT. Children discharged home were contacted one week after their ED visits to assess outcomes, and admitted children were followed through their hospitalizations. Prediction rule test characteristics for ciTBI with 95% confidence intervals (CI) were calculated.

Results:

Results: From December 2016 to December 2020, we enrolled 4,912 children < 2 years old and 12,327 children 2–18 years old. In those < 2 years, 287 (6%) had TBIs on CT and 38 (13%, 95% CI 10, 18%) had ciTBIs. In those 2-18 years, 503 (4%) had TBIs on CT and 143 (28%, 95% CI 25, 33%) had ciTBIs. The prediction rule for those < 2 years old had the following test characteristics: sensitivity 38/38 (100%; 95% CI 92.4 to 100%), specificity 2,531/4,874 (51.9%; 95% CI 50.5 to 53.3%), and negative predictive value (NPV) 2,531/2,531 (100%; 95% CI 99.9 to 100%). The prediction rule for those 2 – 18 years old had the following test characteristics: sensitivity 141/143 (98.6%; 95% CI 95.0 to 99.8%), specificity 5,110/12,188 (41.9%; 95% CI 41.0 to 42.8%), and NPV 5,110/5,112 (99.96%; 95% CI 99.86 to 100%). Neither of the two patients missed by the rule died, had a neurosurgical procedure or were intubated. Conclusion(s):

Conclusion: The PECARN ciTBI rules were externally validated in a large multicenter cohort with excellent test characteristics. Implementation of this rule would further safely decrease CT use.