SGEM#225: NEXUS II – Validation of the Pediatric Head CT Decision Instrument

Subscribe: RSS

[display_podcast]

Date: July 16th, 2018

Reference: Gupta M et al. Validation of the Pediatric NEXUS II Head Computed Tomography Decision Instrument for Selective Imaging of Pediatric Patients with Blunt Head Trauma. AEM July 2018

Guest Skeptics: Dr. Corey Heitz is an emergency physician in Roanoke, Virginia. He is also the CME editor for Academic Emergency Medicine

Case: You’re working in a small rural emergency department when a seven-year-old girl comes in by EMS with a head injury. Her father was teaching her how to bike and he got a little ambitious and sent her down a small hill. She hit a rock, and went over the bars, striking her head on a small tree as she fell. She was helmeted, she did not lose consciousness, has not been vomiting, but the helmet was scratched up where it struck the tree. It’s been one hour since the accident and the child’s exam is otherwise normal; she’s behaving normally and only has a minor headache and some scrapes on her knees. Dad, on the other hand, may need something for anxiety.

Background: Blunt head trauma is a common presenting complaint in emergency departments, accounting for approximately two million visits per year in the US. CT imaging is often performed but comes with radiation risks and increased medical costs.

Several decision instruments have been developed to assess the risk of significant intracranial injury in children with head trauma (CATCH, CHALICE and PECARN). The PECARN tool has been found to have a high sensitivity, but in one study was shown to increase CT use compared to physician judgement.

We covered concussions on SGEM#112. This episode included the pediatric head trauma CT decisions guide for children less than two years of age and those two years of age and older.

The NEXUS Head CT decision instrument was developed as a “one way” instrument, which would hopefully serve to rule out those children who might otherwise receive imaging (as opposed to classifying many as “high risk”.) In the original cohort, use of NEXUS Head CT decision instrument decreased the need for CT by 25%.

Clinical Question: In pediatric patients with blunt head trauma, can the NEXUS Head CT decision instrument be used to rule out the need for imaging in patients who otherwise would have received CT imaging?

Reference: Gupta M et al. Validation of the Pediatric NEXUS II Head Computed Tomography Decision Instrument for Selective Imaging of Pediatric Patients with Blunt Head Trauma. AEM July 2018

Population: Patients less than 18-years-old with blunt head trauma who underwent CT imaging at one of four participating hospitals
Intervention: Clinical judgement followed by the application of the pediatric NEXUS II Head CT decision instrument
Comparison: There is no comparison
Outcome:
- Primary Outcome: Sensitivity, specificity, and negative predictive value (NPV) for the need for neurologic intervention defined as:
  1. Death due to head injury
  2. Need for craniotomy
  3. Elevation of skull fracture
  4. Intubation related to head injury
  5. Intracranial pressure monitoring, within seven days of head injury
Secondary Outcome: Clinically significant head injury on CT imaging
Study Design: Pre-planned secondary analysis of the decision instrument

Dr. William Mower

This is a summer SGEM HOP and we are pleased to have one of the authors on the episode, Dr. William Mower. Bill is a professor in-residence at the UCLA School of Medicine in Los Angeles and among many other things, the director of UCLA Emergency Medicine Research Assistance program.

Pediatric NEXUS II Head CT Decision Instrument?

Pediatric patients with blunt head trauma are classified as low-risk, not requiring CT, if they meet seven criteria:
1. No evidence of skull fracture
2. No scalp hematoma
3. No neurological deficits
4. Normal level of alertness
5. Normal behavior
6. No persistent vomiting
7. No coagulopathy
≥1 positive or unassessed criterion categorizes patient as requiring CT

Authors’ Conclusions: The Pediatric NEXUS Head CT DI reliably identifies blunt trauma patients who require head CT imaging and could significantly reduce the use of CT imaging.

Quality Checklist for Clinical Decision Tools:

The study population included or focused on those in the emergency department. Yes
The patients were representative of those with the problem. Yes
All-important predictor variables and outcomes were explicitly specified. Yes
This is a prospective, multicenter study including a broad spectrum of patients and clinicians (level II). Yes
Clinicians interpret individual predictor variables and score the clinical decision rule reliably and accurately. Unsure
This is an impact analysis of a previously validated CDR (level I). No
For Level I studies, impact on clinician behavior and patient-centric outcomes is reported n/a
The follow-up was sufficiently long and complete. Unsure
The effect was large enough and precise enough to be clinically significant. Yes

Key Results: The original NEXUS CT head validation observational study had close to 8,000 patients with blunt head injury. There were 1,018 patients less than 18 years old who received head CT scans. This cohort included 27 patients (2.7%) who required neurological intervention, and 49 patients (4.8%) had significant intracranial injuries.

All 27 patients requiring neurosurgical intervention were identified by the Pediatric NEXUS II Head CT Decision Instrument.

Primary Outcome: Need for neurosurgical intervention
- Sensitivity: 100% (95% CI 87.2%–100%) – 27 of 27
- Specificity: 33% (95% CI 30.3-36.3%) – 330 patients of 991 who did not require intervention were classified as low-risk status by the Pediatric NEXUS II Head CT DI
- Negative Predictive Value: 100% (95% CI = 99.6%–100%) – None of the 991 low-risk patients required neurosurgical intervention

Secondary Outcome: Clinically significant head injury
- Sensitivity: 98% (95% CI 89.1%– 99.9%) – 48 of 49 patients with significant injury were identified by the Pediatric NEXUS II Head CT DI
- Specificity: 34% (95% CI 31.0%– 37.0%) – 329 of 969 patients who did not have significant injury were classified as low risk by the Pediatric NEXUS Head CT DI
- Negative Predictive Value: 99.7% (95% CI 98.3%–100%) – 329 of 330 low-risk patients were absent of a clinically significant head injury

Listen to the podcast on iTunes to hear Bill’s responses to our ten (two sets of five) nerdy questions.

Secondary Analysis: This was a pre-planned secondary analysis of the NEXUS Head CT decision instrument. Do you think this weakens or limits the conclusions of your study?
Power: The original NEXUS Head CT observational study was powered to have 368 patients with injuries requiring neurosurgical intervention. It was not powered for the subgroup of pediatric patients reported in this study. There were only 27 pediatric patients who required intervention in the cohort. Can you comment on the power of your study to detect the primary outcome of need for neurosurgical intervention?
Spectrum Bias/Verification Bias: The original study did not enroll patients that did not have a CT scan performed. This can introduce spectrum bias and verification bias. How did you address this issue?
Exclusions: You state that clinically significant intracranial injury excluded some intracranial hemorrhage, some skull fractures, and isolated pneumocephaly. In my experience, it’s standard practice that if these injuries are found on CT, the vast majority of these patients are admitted. Was any data kept as to how many patients had one of these “nonsignificant” injuries, either in the CT cohort or the verification bias cohort?
Intra-Rater Reliability: Clinical decision tools rely heavily on the ability of the individual components to be evaluated consistently by different clinicians. Can you tell us how consistent and accurate clinicians are at assessing the components of the Pediatric NEXUS II Head CT Decision Instrument?
One Miss: There was one patient with significant injury who was misclassified. Can you talk about that patient a little? Is this a case where inconsistency in assessing criteria may have played a role?
Confidence Interval: The 95% confidence interval around a point estimate gives the precision of results. Your 95% confidence interval around the primary outcome of need for neurosurgical intervention was wide (87% – 100%). This lack of precision and lower boundary of 87% is a concern given the serious patient-oriented outcome. What do you think SGEMers should do with this data?
Follow-up: Do you think the seven-day follow-up was long enough to capture all the possible significant injuries?
One-Way Tool: NEXUS is a “one-way” clinical decision instrument. Can you discuss this concept of a “one-way” tool and how NEXUS is different from PECARN in that regard?
Anything Else: Those are the ten nerdy questions. Is there anything else you would like to tell the SGEMers about your study?

Comment on Authors’ Conclusion Compared to SGEM Conclusion: We agree with the authors’ conclusions.

SGEM Bottom Line: The Pediatric NEXUS II Head CT Decision Instrument can reliably categorize patients as low risk and may reducing CT imaging in these patients.

Case Resolution: In your patient, all of the low-risk features are present. You discuss with the father that in your clinical judgement you do not feel a head CT is indicated and discuss observation precautions and symptoms to look for and dad seems relieved.

Dr. Corey Heitz

Clinical Application: A properly powered study with tighter confidence intervals around the point estimate for the patient-oriented outcome of need for neurosurgical intervention is desireable before the Pediatric NEXUS II Head CT Decision Instrument is widely adopted.

What Do I Tell My Patient? I tell the father that imaging isn’t indicated given the extremely low risk of significant injury. I commend the patient for wearing her helmet and instruct the father that any time a helmet is involved in an accident the recommendation is to replace it.

Keener Kotest: The last winner was Liam Portt a medical student from Sterling, Ontario. He knew it was in 1957 when a French Physician, Dr. Paul Nogier, created an ear acupuncture map.

Listen to the SGEM podcast on iTunes to hear this weeks’ question. If you know the answer send an email to TheSGEM@gmail.com with “keener” in the subject line. The first correct answer will receive a cool skeptical prize.

SGEMHOP: Now it is your turn SGEMers. What do you think of this episode? Tweet your comments using #SGEMHOP. What questions do you have for Dr. Mower and his team about getting head CTs in pediatric patients with blunt head trauma? Ask them on the SGEM blog. The best social media feedback will be published in AEM.

Also, don’t forget those of you who are subscribers to Academic Emergency Medicine can head over to the AEM home page to get CME credit for this podcast and article. We will put the process on the SGEM blog: