PECARN Head CT Clinical Decision Rule
The Study: Identification of children at very low risk of clinically important brain injuries after head trauma: a prospective cohort study. Kuppermann, N. et al. The Lancet. 2009 Oct;324:1160-70.
Bottom line: The two validated prediction rules borne of this study adequately identify children at low risk of clinically important traumatic brain injury (ciTBI), and children meeting these criteria should not undergo head CT for their head injury.
The idea: TBI is the leading cause of death and disability in children in the world. Head CT is used to evaluate patients with head injury because ciTBI needs to be identified rapidly for intervention (e.g. neurosurgery). Prior to this study, the authors found that about 50% of children with head injuries were undergoing head CTs in the course of their workups. In particular, children with GCS 14-15 with minor head trauma are frequently scanned; however, less than 10% of these CTs show brain injury, with even fewer cases requiring neurosurgery. CT scanning is not without its risks; it’s estimated that 1 in 1000 to 1 in 5000 pediatric head CTs will lead to a lethal malignancy, and many children need to be sedated for imaging which carries its own risks. The purpose of this study was to develop a clinical decision rule to determine which children are at low risk of ciTBI and in which cases CT can be avoided.
Methods: This was a prospective cohort study of children less than 18 years old presenting to 25 EDs across North America with acute head trauma with a GCS of 14 or 15. The study first enrolled a derivation population to derive the clinical decision rule followed by a validation population. Exclusion criteria included delayed presentation (>24 hour), trivial mechanisms of injuries (e.g. ground level falls), patients with penetrating trauma, known brain tumors or neurologic disorders, VP shunts, or bleeding disorders. Children were assessed for ciTBI defined as death from TBI, neurosurgery, intubation >24h for TBI, or hospital admission of 2 or more nights associated with TBI seen on CT. Children <2 years old and children 2 years or older were analyzed separately given differences in communication ability, radiation sensitivity, and different risk profiles for TBI.
Results: 42412 patients were included in this study: 33785 in the derivation population and 8627 in the validation population. 376 (0.9%) had ciTBIs.
In children less than 2 years old, children that had none of the following risk factors were at very low risk of ciTBI:
Altered mental status
Non-frontal scalp hematoma
Loss of consciousness > or equal to 5 seconds
Severe mechanism of injury defined by MVC with patient ejection, rollover, death of another passenger, pedestrians or bicyclists without helmet struck by vehicle, fall from >3 feet, or head struck by high impact object
Palpable skull fracture
Not acting normally according to parent
CT scans were obtained in approximately 31% of patients in this age group, 25% of which were in children that met low risk criteria for ciTBI. The negative predictive value of the decision rule in the validation population was 100% (1175/1175) with a sensitivity of 100% (25/25) for ciTBI when patients had at least one of the risk factors.
In children 2 years or older old, children that had none of the following risk factors were at very low risk of ciTBI:
Altered mental status
Loss of consciousness
History of vomiting
Severe mechanism of injury defined by MVC with patient ejection, rollover, death of another passenger, pedestrians or bicyclists without helmet struck by vehicle, fall from >5 feet, or head struck by high impact object
Clinical signs of a basilar skull fracture
37% and 34.7% of patients in this age group had CT scans in the derivation and validation studies respectively. 20.7% and 19.7% were in children that met low risk criteria for ciTBI. The negative predictive value of the decision rule in the validation population was 99.95% (3696/3698) with a sensitivity of 96.8% (61/63) for ciTBI in patients that had at least one of the risk factors. In the two children that were misclassified by the decision rule in the validation population, neither required neurosurgery.
Discussion: The strengths of this study are numerous. The decision rules were not only derived in this study but also validated in a separate population. The population used was large and diverse. The clinical decision rules are age group specific to account for important clinical differences in these groups. The rules are simple, intuitive, easy to assess, and have a high negative predictive value, and show that approximately 20-25% of CTs performed in children in this study could have been avoided with minimal consequences. This benefit may be even higher at non-pediatric EDs where providers are less comfortable taking care of children with head injuries.
Conclusion: The authors suggest an algorithm that recommends CT imaging for children displaying at least one of the two most predictive indicators for each age group (altered mental status and palpable skull fracture in children <2, altered mental status and signs of a basilar skull fracture in children 2 years and older), no CT imaging if none of the 6 indicators are present, and CT imaging vs. observation based on clinical judgment and parental preference for presence of one or more of the 4 less predictive indicators in each age group.