Imaging/Concepts
Developing a clinical decision instrument to rule out intracranial injuries in patients with minor head trauma: Methodology of the NEXUS II investigation*,**,*

https://doi.org/10.1067/mem.2002.129245Get rights and content

Abstract

The fear of failing to identify brain injury has led to the liberal and potentially excessive use of computed tomographic (CT) scanning of patients with blunt head trauma who have even a remote possibility of intracranial injury. This practice exposes large numbers of patients to the expense and radiation exposure associated with CT imaging while detecting injuries in a small minority. Previous studies suggest that it might be possible to develop a decision instrument to identify patients with blunt head injury who have essentially no risk of significant intracranial injury and for whom CT scanning is therefore unnecessary. Development of such a decision instrument has been identified as a priority among practicing emergency physicians. The National Emergency X-Radiography Utilization Study II (NEXUS II) is a large, multicenter, prospective study designed to derive a decision rule for CT imaging of patients with blunt head injury. This study, conducted in 21 different emergency departments across the United States and Canada, will enroll more than 10 times as many patients with head trauma as any currently published study. NEXUS II should be able to definitively answer questions about the validity and reliability of clinical criteria as a preliminary screen for blunt head trauma. [Ann Emerg Med. 2002;40:505-514.]

Introduction

Unrecognized intracranial injury (ICI) can produce permanent brain damage, severe disability, and even death. Sporadic reports of occult injuries have generated concern, and many clinicians have adopted a policy of obtaining computed tomographic (CT) scans on virtually all patients with blunt head trauma. Clinical reports suggest that approximately 1 million patients with blunt trauma undergo head CT scanning in the United States each year.1, 2, 3 These studies generate nearly $750 million in charges but reveal significant ICI in less than 60,000 patients. Thus, the vast majority of patients who undergo CT (approximately 95%) incur the expense and radiation exposure of a negative examination result.1, 4, 5, 6

Recent work suggests that it might be possible to develop a decision instrument by using a limited number of clinical criteria accessible to the treating physician in the emergency department to identify patients with blunt head injury who have essentially no risk of brain injury.7, 8 Development of a decision instrument that can accurately identify patients who have no need for head CT imaging is a priority among practicing emergency physicians.9, 10, 11 Such an instrument might reduce head CT imaging by as much as 20%, with a resultant $150 million decrease in annual radiographic charges and significant reductions in radiation exposure and the attendant risk of radiation-induced thyroid and other malignancies.

In this 2-part multicenter prospective study, all patients with blunt trauma presenting to a participating ED will undergo standardized clinical evaluation before CT imaging. In the first phase of the study, the interrater variability of candidate variables will be assessed to identify criteria that can be reliably measured in emergency settings. Any and all such criteria will then be used in the derivation phase. The presence or absence of each reliable criterion, as well as the presence or absence of ICI, will be documented for each imaged patient. Recursive partitioning will be used to derive a decision instrument that identifies a no-risk subset of patients who do not have significant ICI. The final instrument will be expected to exhibit a sensitivity and negative predictive value (NPV) of 100% for ICI while retaining the highest possible specificity.

Section snippets

Methods

NEXUS II is a multicenter, prospective, observational study of ED patients with blunt trauma for whom head CT scanning is ordered. Participating centers represent a wide variety of facilities, including university hospitals, community hospitals with and without teaching programs, public hospitals and private hospitals, and hospitals with all levels of trauma categorization. Study sites are organized into geographic regions distributed across the country. The wide range of hospitals serves to

Discussion

The current liberal use of CT in patients with blunt head trauma stems from the concern that significant ICI can occur even among patients sustaining minor head trauma. Fear of missing such an ICI, with its potential to produce severe neurologic disability or even death, has led some authors to advocate CT imaging in virtually all patients with blunt head injury. This strategy generates large numbers of negative CT scans for each injury detected. Furthermore, many of the injuries detected do

Internal validity

The study's internal validity could be weakened by errors in estimating the criteria's sensitivities and NPVs. Bias could occur if a patient's injury status was misclassified or if physicians change their assessments after reviewing CT scans. Bias caused by altered assessments will be eliminated by requiring data to be collected and entered before generation of CT imaging requests. This leaves the potential for misrepresentation of critically ill patients, but these patients will, in general,

References (21)

There are more references available in the full text version of this article.

Cited by (52)

  • Physical Examination Sensitivity for Skull Fracture in Pediatric Patients With Blunt Head Trauma: A Secondary Analysis of the National Emergency X-Radiography Utilization Study II Head Computed Tomography Validation Study

    2023, Annals of Emergency Medicine
    Citation Excerpt :

    Overall, this suggests that providers evaluating pediatric patients presenting with blunt head trauma may miss a substantial number of skull fractures during their initial assessment. Despite numerous studies indicating that examination evidence of skull fracture is predictive of intracranial injury, we are aware of few studies investigating provider ability to detect these injuries.8,12,13,16,17 Interestingly, an evaluation by Tunik et al21 of basilar skull fractures in a Pediatric Emergency Care Applied Research Network cohort indicated provider examination missed approximately 60% of these injuries.

  • Prevalence of Intracranial Injury in Adult Patients With Blunt Head Trauma With and Without Anticoagulant or Antiplatelet Use

    2020, Annals of Emergency Medicine
    Citation Excerpt :

    The decision to obtain CT imaging was based on the clinical judgment of the treating physician and was not dictated by study protocol. Other methodological details have been published previously.20 Research assistants were trained to approach the treating clinicians and collect demographic, clinical, and medication information on each patient, using a standardized data collection form before CT (Figure E1, available online at http://www.annemergmed.com).

View all citing articles on Scopus
*

A condensed version of this article appears in the print journal.

**

Supported by grant RO1 HS09699 from the Agency for Healthcare Research and Quality.

*

Address for reprints: William R. Mower, MD, PhD, UCLA–EMC, 924 Westwood Boulevard, Suite 300, Los Angeles, CA 90024; fax 310-794-0599; E-mail [email protected]

View full text