The RAMER Reviews: PECARN prediction rule for cervical spine imaging of children presenting to the emergency department with blunt trauma: a multicentre prospective observational study
Written By: Sabrina Rosengarten, MD, MPH; Edited by: Timothy Khowong, MD, MSEd
The Study:
This study, published in Lancet in July, 2024, aimed to create a new tool to safely avoid unnecessary CTs and XRs in children with potential c-spine injuries. Research has shown that a single cervical spine CT increases the risk of thyroid cancer by 78%. Although with technology advances in CT scanners the radiation dose has been reduced over time, it is still worth considering reducing exposure as much as possible. It additionally can save a significant amount of time for patients, physicians, nurses and technicians if a CT can be avoided. As is well known by clinicians, trying to get a CT on a 2 year old can be a very difficult procedure, sometimes even requiring procedural sedation.
This was a prospective observational study performed at 18 different children’s hospitals across the country. They split the hospitals into a derivation cohort and a validation cohort. They administered questionnaires to the ED attending that asked for their thoughts on risk factors that the patient had for cervical spine injury prior to seeing imaging results. Later, the chart was reviewed and the patient was called within 28 days of their visit to the ED to see if they had been diagnosed with a cervical spine injury.
Patients were deemed eligible for the study if they were transported from the scene of injury by EMS, or if they underwent cervical spine imaging at the participating or transferring hospital. Children with penetrating trauma were excluded. This was a large study with over 22,000 participants, so it had sufficient power to answer their research question. They used a special regression analysis to determine which risk factors were associated with a non-negligible risk of cervical spine injury. They found that GCS/AVPU score, abnormal airway, breathing or circulation, and focal neurological deficits on exam were all correlated with a higher risk of c-spine injury. These were all statistically and clinically significant. AMS, neck pain or neck tenderness, and substantial head or torso injury were also correlated with a higher risk of c-spine injury based on their analysis, but at a lower incidence. Using those findings, they then created the scoring tool and applied it to the validation cohort.
Limitations of the study include that it was performed at children’s hospitals, so it may be less applicable to community sites. Additionally, the tool may be best applied differently at different ages, but all ages were grouped together. And finally, there are still subjective components of the tool that leave room for clinical gestalt or human error. It appeared to be internally valid, except for one noted statistical difference between the validation and derivation cohorts, which was that children in the validation cohort were less frequently admitted to the ICU or OR than those in the derivation cohort.
This tool was very well made and can definitely positively inform practice. As they estimated, the number of children having CT might have decreased from 17.2% to 6.9% in their study population if they had used the score from the beginning. It is from a reliable source, and statistically reliable. I will certainly be using this tool to help inform my imaging decisions in the future.
Below is the basic scoring tool that was created.
Reference:
Leonard, J. C., Harding, M., Cook, L. J., Leonard, J. R., Adelgais, K. M., Ahmad, F. A., Browne, L. R., Burger, R. K., Chaudhari, P. P., Corwin, D. J., Glomb, N. W., Lee, L. K., Owusu-Ansah, S., Riney, L. C., Rogers, A. J., Rubalcava, D. M., Sapien, R. E., Szadkowski, M. A., Tzimenatos, L., . . . Kuppermann, N. (2024). PECARN prediction rule for cervical spine imaging of children presenting to the emergency department with blunt trauma: a multicentre prospective observational study. The Lancet Child & Adolescent Health, 8(7), 482–490. https://doi.org/10.1016/s2352-4642(24)00104-4