In the recent National Emergency X-Radiography Utilization Study (NEXUS) II, a decision rule with eight clinical criteria-evidence of significant skull fracture, altered level of alertness, neurological deficit, persistent vomiting, presence of scalp hematoma, abnormal behavior, coagulopathy and advanced age-was developed to rule out intracranial injury (ICI) in patients with blunt head trauma, with the intention of obviating the need for CT scanning.OBJECTIVE
To investigate the performance of the NEXUS II decision rule in pediatric patients with blunt head trauma.DESIGN AND INTERVENTION
The present analysis included all pediatric patients (aged ≤18 years) who participated in NEXUS II-a prospective, observational study of patients with blunt head trauma who were enrolled at 21 emergency departments across North America; the study period was May 1999 to December 2000. All patients had a cranial CT as part of their emergency department evaluation in order to identify clinically important ICI, defined as ICI that required neurosurgical intervention or was likely to be associated with significant long-term neurological impairment. The performance of the NEXUS II decision instrument (excluding the criterion of advanced age) was evaluated in the whole pediatric cohort (n=1,666) and in children who were younger than 3 years (n=309).OUTCOME MEASURES
Sensitivity, specificity and negative predictive value (NPV) of the NEXUS II instrument.RESULTS
In total, 138 children (8.3%) had clinically significant ICI. Clinicians were able to evaluate most variables included in the NEXUS II instrument in the majority of these patients; neurological deficit (including Glasgow Coma Scale score <15), altered level of alertness and abnormal behavior were reported as the most frequent findings. The NEXUS II decision rule correctly identified 136 of the 138 ICI cases and classified 230 cases as low risk, resulting in a sensitivity of 98.6% (95% CI 94.9-99.8%), a specificity of 15.1% (95% CI 13.3-16.9%) and an NPV of 99.1% (95% CI 96.9-99.9%). The two children not identified with the decision instrument did not require neurosurgical intervention. The NEXUS II instrument performed even better in patients who were aged <3 years, 25 of whom (8.1%) had clinically significant ICI. All of these cases were correctly identified using the decision rule (sensitivity 100%, 95% CI 86.3-100%), and 15 cases were classified as low risk (specificity 5.3%, 95% CI 3.0-8.6%; NPV 100%, 95% CI 78.2-100%). The most common NEXUS II criterion observed in children under the age of 3 years with ICI was altered level of alertness, followed by scalp hematoma. Physicians were unable to assess neurological function completely in eight of these children, but all children under the age of 3 years with ICI had at least one of the other risk factors included in the NEXUS II instrument.CONCLUSION
CT imaging in children with head injury can be limited to those who exhibit at least one of the amended NEXUS II risk criteria.