One common and vexing problem I’ve run into thus far in residency is the intoxicated patient, found down, brought in by EMS in a rigid cervical collar placed because of the presumption of possible trauma leading to an unstable cervical injury. The efficacy and necessity of cervical collars has been debated elsewhere, and I’m not looking to discuss that here — what I’m more interested is, if these patients have a negative CT scan (for better and for worse, fairly common practice in those unable to give a reliable exam, especially if they have any sign of trauma on them), can we safely remove their collar?
This study, by the “Pacific Coast Surgery Association” and published in JAMA Surgery, prospectively evaluated 1668 intoxicated adults with blunt trauma who underwent cervical spine CT scans over one year at a single Level I trauma center. Intoxication was defined based on the results of urine and blood testing, and the outcome of interest was clinically-significant cervical spine injuries that required cervical immobilization (not necessarily surgical fixation).
The authors wanted to evaluate the negative predictive value of a normal CT scan in the intoxicated patient to determine whether this would allow safe removal of their cervical collar– it is well-known that some injuries (e.g. unstable ligamentous injuries or spinal cord injuries without fractures of the vertebrae) may not be identifiable on a CT scan, and in the patient who is altered, it may be difficult to elicit exam findings that would tip a practitioner off to the presence of these injuries.
So what did they find? In intoxicated patients, the negative predictive values of a CT scan read as negative for acute injury were 99.2% for all injuries and 99.8% for unstable injuries. There were five false-negative CTs, with 4 central cord syndromes without associated fracture. There was also one false-negative for a potentially unstable injury identified in a drug-intoxicated patient who presented with clear quadriplegia on examination. All of these were detected on MR imaging. About half of the intoxicated patients with the negative CT went on to be admitted with their cervical collar left on. None of these intoxicated patients went on to have an injury identified later, or to have any neurologic deficit, leading to a conclusion of a NPV of 100% in that cohort.
My takeaway from this paper: while there are some weaknesses, e.g. the lack of protocol-based care and the significant heterogeneity in terms of “intoxication”, it seems reasonable to take away from this that a negative CT scan done on a modern scanner and read by an experienced trauma radiologist or neuroradiologist does allow you to safely clear the collar of an intoxicated patient who does not have any gross neurologic deficits. This data lends further support to the 2015 recommendations from the Eastern Association for the Surgery of Trauma who in a systematic review and meta-analysis “found the negative predictive value for identifying unstable CSIs to be 100% and thus have made a conditional recommendation for cervical collar removal based on a normal high-quality CT scan”. Adopting this practice could help minimize unnecessary testing (including expensive MRIs that are more likely to show false positives than to identify clinically-significant injuries) , allow for earlier disposition of patients from the emergency department, increase patient comfort, and decrease the emotional and cognitive burden placed on providers who otherwise often have to continuously struggle to keep patients adherent to immobilization practices.