Written by Clay Smith
Paired head and c-spine CT scans had low yield for detecting clinically significant injury (CSI) on both, 0.5%. This suggests a selective scanning, rather than a shotgun approach, may be better.
Why does this matter?
We often order a head CT along with a c-spine CT and vice versa. But often there is only clinical suspicion of injury to one area. The cost and radiation exposure of each scan is not trivial. How often are serious injuries found on both?
This was a retrospective, single center study of over 3000 patients who had c-spine CT ordered as part of the trauma workup. Specifically, they wanted to know how many had a head CT paired with the c-spine CT. They were paired >89% of the time, N = 2888. There were 251-274 (8.7 - 9.5%) patients with a priori defined clinically significant injury (CSI) to the head. Three different numbers were given in the text and tables, so I don't know which is correct. For c-spine 70-98 (2.4 - 3.4%) had CSI to the c-spine. Again, 3 different numbers were reported. It's pretty frustrating this wasn't caught by the authors or in peer review. The proportion with CSI to the head and c-spine was 0.5% (14/2888), and for any injury (not just "clinically significant"), 1.4%. The take home point is we don't have to order both unless there is clinical suspicion to warrant the scans. The only exception, in my mind, is in the elderly. I have been surprised to find bad c-spine fractures in elderly patients with no or minimal symptoms, but this is just my personal, anecdotal experience, not evidence.
Low Yield of Paired Head and Cervical Spine Computed Tomography in Blunt Trauma Evaluation. J Emerg Med. 2018 Jun;54(6):749-756. doi: 10.1016/j.jemermed.2018.02.015. Epub 2018 Apr 22.
Open in Read by QxMD
Reviewed by Thomas Davis