The c-spine clearance protocol at most hospitals requires that patients with c-spine tenderness get either a flexion-extension c-spine x-ray or MRI with STIR sequence, even in the face of a negative c-spine CT (98% sensitivity by some studies). Alternatively, they can simply stay in a c-collar and follow up in a couple weeks to see if they can be cleared clinically at that time. "Unreliable" patients, however (those who are intoxicated, unconscious, too young to communicate, have distracting injuries or are otherwise uncooperative), cannot get flex-ex x-rays. Thus, the MRI ends up being the tool of choice for clearing these folks' c-spines.
At many institutions, the protocol is to get the MRI within 48 hours, and this protocol has developed into a powerful dogma, with many claiming that the STIR sequence is unreliable for the identification of ligamentous injury/edema unless done within 48 hours. Anything after 48 hours is considered unreliable, and the patient is thus bound to a c-collar until they can be cleared clinically or become capable of flex-ex. (Some centers will also do dynamic flex-ex under fluoroscopy in unconscious patients, but this is less common and very staff intensive.)
Is the STIR MRI really unreliable after 48 hours? I've searched for an answer to this question many times and come up empty-handed. I finally asked a very evidence-based neurosurgeon I know, and he gave me this article:
Radiographic Assessment of the Cervical Spine in Symptomatic Trauma Patients.
GUIDELINES FOR THE MANAGEMENT OF ACUTE CERVICAL SPINE AND SPINAL CORD INJURIES
Neurosurgery. 50(3) Supplement:S36-S43, March 2002.These guidelines say that c-spine can be cleared following "a normal MRI study obtained within 48 hours of injury," but the article contains only one line "explaining" why:
MRI studies are preferred within the first 48 hours after injury.
It doesn't say why, it only gives these references:
Benzel E, Hart B, Ball P, Baldwin N, Orrison W, Espinosa M.
Magnetic resonance imaging for the evaluation of patients with occult cervical spine injury.
Journal of Neurosurgery 85: 824-829, 1996.
D'Alise M, Benzel E, Hart B.
Magnetic resonance imaging evaluation of the cervical spine in the comatose or obtunded trauma patient.
Journal of Neurosurgery Spine 91(Spine 1):54-59, 1999.
Emery S, Pathria M, Wilber R, Masaryk T, Bohlman H.
Magnetic resonance imaging of posttraumatic spinal ligament injury.
Journal of Spinal Disorders 2: 229-233, 1989.
Katzberg R, Benedetti P, Drake C, Ivanovic M, Levine R, Beatty C, Nemzek W, McFall R, Ontell F, Bishop D, Poirier V, Chong B.
Acute cervical spine injuries: prospective MR imaging assessment at a level one trauma center.
Radiology 213: 203-212, 1999.
White P, Seymour R, Powell N.
MRI assessment of the pre-vertebral soft tissues in acute cervical spine trauma.
British Journal of Radiology : 818-823, 1999.I read every single one of these articles, and NOT ONE of them offered ANY evidence in favor of the 48-hour rule. They simply stated as part of their protocol that scans were obtained within X number of hours from time of injury. The second Benzel article states (without any evidence cited):
In our experience the ability to detect soft-tissue injury diminishes after 72 hours.
The Emery article, in my opinion, is most telling: they evaluated 37 patients with c-spine injuries:
Nineteen patients were considered to have torn posterior ligaments on the basis of their surgical, clinical, and/or radiographic findings. MRI detected ligament damage in 17 of these 19 cases (Figs. 1-4). One case of ligament injury not identified by MRI was in the cervical spine of a 23-year-old male that was imaged 40 days after injury.
The other missed injury was imaged 1 day after injury, and was performed without T2 sequence at another hospital. The average timing of the MRI scan in this study was
10.8 days after injury, and yet 17 of 19 ligamentous injuries were still identified. Furthermore, this study was conducted in 1989 on either 1T or 1.5T scanners. It would be absurd to suggest that MRI technology had not advanced in the last 19 years. The use of 3T scanners is now commonplace, and the software algorithms have improved tremendously. This article did not even make mention of STIR sequence, and I'm not sure that it even existed back then.
Ligamentous injuries can clearly be identified at significantly greater than 48 hours. So, why the 48 hour rule? I spoke with one of my attendings at GWU about this, and the answer I got did, in fact, make sense: getting the collar off has numerous advantages, including improving venous drainage, decreasing aspiration risk, facilitating procedures, increasing patient comfort and decreasing the risk of ulcer development. It is thus advantageous to get it done as soon as possible, and it is true that edema does resolve with time. How much time is not clear, but you might as well set yourself a goal for imaging, and 48 hours is as good a goal as any.
Thus, it is certainly reasonable to attempt to get MRI images of unreliable, symptomatic patients with potential c-spine injuries within 48 hours, but it would appear to be simply untrue that the images will be unreliable or in any way inadequate after the 48 hour period.
-- David
