Spinal Radiography: State-Of-The-Art
By Michael A. Gibbs, MD, FACEP
An essential part of early trauma care involves the assessment and management of the patient with the potential for spinal injury. This potentially devastating injury can have lifelong implications for the patient, and if unrecognized or poorly managed, for the treating physician as well. As specialists in emergency medicine, it is not good enough for us to be "familiar" with the management of these high-risk patients; we must become experts. This requires a thorough understanding of the biomechanics of trauma, refined clinical examination skills, an up-to-date knowledge of the indications for spinal radiography, and a sophisticated approach to the interpretation of these x-rays. This feature will outline the current indications for cervical spine radiography. In addition, I will review a series of articles published by the NEXUS (National Emergency X-ray Utilization Study) investigators in the July 2001 issue of the Annals of Emergency Medicine.
Lowery DW, et al. Epidemiology of cervical spine injury victims. Ann Emerg Med 2001;38:12-15. This paper was a standard epidemiologic review of the NEXUS database (n = 34,069, 818 [2.4%] with injury); 71% of subjects were male and the incidence of spinal injury was greatest in young adult males, ages 15-35. The prevalence of spinal injury increased progressively with age.
Message: Recognize that young patients are the most likely to sustain cervical spine fracture, but also keep in mind that elderly patients are especially vulnerable.
Goldberg W, et al. Distribution and patterns of blunt traumatic cervical spine injury. Ann Emerg Med 2001;38: 17-21. The NEXUS database was reviewed to describe the level and location of cervical injury. The second cervical vertebra was the most common level of injury (n = 286 [24%]). Four hundred-seventy fractures (39%) occurred at C6 and C7.
Message: While cervical fractures occur at all levels, the upper and lower vertebrae are the most likely to be injured. This is especially relevant because these are the two most difficult regions to define radiographically. Physicians should pay close attention to these.
Indications For Radiography
The indications for cervical spine radiography have evolved considerably during the past 20 years. In the 1980s the American College of Surgeons recommended cervical radiography in "any patient with major blunt trauma." While this approach was effective for excluding injury, it came at the expense of a large number of negative radiographs. In addition to the obvious financial implications, this approach resulted in unnecessary exposure to radiation, and potential delays in trauma patient evaluation and stabilization. In the decade that followed, several authors attempted to identify clinical criteria that could reliably exclude cervical injury. While these studies suggested that low-risk criteria could successfully be employed, none had the statistical power to support their conclusions.
These efforts culminated in the National Emergency X-radiography Utilization Study (NEXUS).1 NEXUS was a prospective, multicenter, observational study of a decision rule used to identify patients at low risk of cervical spine injury and, thus, not requiring cervical radiography. The decision instrument required patients to meet five criteria in order to be classified as having a low probability of injury: 1) no midline cervical tenderness; 2) no focal neurologic deficit; 3) normal alertness; 4) no intoxication; and 5) no painful, distracting injury. More than 34,000 patients at 21 academic and non-academic medical centers were evaluated. Physicians were asked to assess each of the clinical criteria before radiographs were available. No efforts were made to influence whether physicians ordered radiographs; these were obtained at the discretion of the treating clinician. The decision rule identified all but eight of 818 patients who had cervical spine injury (sensitivity 99.0% [95% CI 90.0-99.6%]; negative predictive value 99.8% [95% CI 99.6-100%], specificity 12.9%, positive predictive value 2.7%). Only two of the patients classified as unlikely to have an injury according to the decision rule met the preset definition of a clinically significant injury, and one of these two patients required surgical stabilization. Using the decision instrument, radiographic imaging could have been avoided in 4309 (13%) of the 34,069 patients. The inter-rater reliability of these criteria is substantial, although it should be remembered that they may be subjective.2 The assessment of a "distracting" injury is particularly subjective and problematic. A recent article by the NEXUS study group sites that a significant fraction of blunt trauma patients may have an injury considered to be distracting.3 Ullrich and colleagues prospectively evaluated 778 patients and found that 264 (34%) had distracting, painful injuries (DPIs). Fractures accounted for the majority of DPIs (154 or 58%); 42 (16%) were soft-tissue injuries or lacerations, and 86 (34%) were due to a variety of other entities, including visceral, crush, burn, or other miscellaneous injuries. Among the 37 (5%) patients with cervical fractures, 20 (54%) had a DPI, including three (8%) who had a DPI as the only indication for cervical radiography. Clinical judgment should guide practice in these situations.
Panacek EA, et al. Test performance of the individual NEXUS low-risk clinical screening criteria for cervical spinal injury. Ann Emerg Med 2001;38:22-25. Secondary analysis of the NEXUS database was performed to determine the contribution of each of the five individual criteria to the overall sensitivity of the decision instrument. In patients with injury, no one criterion was found in the majority of patients. Fifty percent of patients had midline tenderness; 30% had only one criterion, and one-half of these had only midline tenderness.
Message: Don’t cut corners or hang your hat on any one criterion. It is interesting to note that a significant number of patients with injury and only one criterion did not have midline tenderness.
Which X-Rays Should Be Obtained?
Is the Cross Table Lateral View Sufficient? No! It has been demonstrated that the use of a cross table lateral view alone is inadequate to rule-out cervical spine injury, with a sensitivity of between 57% and 85%. MacDonald and colleagues found that the addition of the axial projection (AP) and odontoid views to the cross table lateral increased the sensitivity from 83% to 99%.4 For this reason, at least three views should be obtained in all cases.
It has been argued that a cross table lateral may be useful to "clear" the cervical spine prior to endotracheal intubation. This approach has several important pitfalls: 1) it may delay definitive airway management; 2) it does not exclude cervical injury, and thus may provide the operator with a false sense of security; and 3) there is no evidence to support an increased risk of neurologic injury during endotracheal intubation, provided in-line stabilization and effective immobilization are maintained at all times.
Three Views or Five? The issue of whether oblique views are routinely needed remains controversial. These views are held by some to be essential because they provide superior visualization of the posterior column (pedicles, articular pillars, neural foramina, and lamina). Turetsky and colleagues found that the oblique views demonstrated certain fractures that were not detected on the three-view series.5 Conversely, Freemyer and colleagues found no fractures or dislocations detected on the five-view series that were not identified on the three-view series.6 While there is no consensus concerning the necessity for routine oblique radiographs in cervical trauma, these views may be useful for evaluating poorly visualized areas of the posterior column. In addition, the supine oblique view provides excellent definition of the cervicothoracic junction, and may be used instead of the often-inadequate swimmer’s view.7,8 Ireland and colleagues compared 60 patients whose cervical spines were imaged with swimmer’s views to evaluate the cervicothoracic junction to those of 62 patients whose junctions were imaged with bilateral supine oblique radiographs.9 Oblique views identified the junction adequately in 38%, compared to 37% in the swimmer’s group. However, the facet joints and posterior elements were fully interpretable in 70% of those imaged obliquely, compared to only 37% in the swimmers group. It is reasonable to use the oblique view selectively, after a three-view series has been evaluated.
Mower WR, et al. Use of plain radiography to screen for cervical spine injuries. Ann Emerg Med 2001;38:1-7. This was a review of the NEXUS database to document the efficacy of plain film radiography and to categorize the frequency and type of injuries missed. A "standard" three-view series was obtained in all patients, with additional imaging studies left to physician discretion. Two hundred thirty-seven patients (0.67% of total, 29% of injury group) with inadequate films had missed injuries. Twenty-three patients with adequate films had missed injuries (0.069% of total, 2.8% of injury group); three of these (0.36% of injury group) were unstable.
Message: In patients with adequate radiographs, the number of significant missed injuries is small, although not zero. More importantly, settling for inadequate radiographs, which can be anticipated in up to one-third of patients, is a recipe for disaster.
What is the Role of Flexion-Extension (F/E) Views? Neurologically intact patients with persistent neck pain and tenderness despite normal radiographs should have F/E views performed to exclude ligamentous injury. It is essential that the patient be alert and cooperative, as all neck movement must be patient-initiated and discontinued immediately should pain occur. Manipulation of the neck by the physician or radiology technologist to overcome spasm is absolutely contraindicated. Filming in the erect position is preferred because this position better demonstrates ligamentous instability.10
Pollack CV, et al. Use of F/E radiographs of the cervical spine in blunt trauma. Ann Emerg Med 2001;38:8-11. Review of the NEXUS database was performed to describe the contribution of F/E films to radiographic evaluation. Of 818 patients with cervical injury, 86 (10.5%) underwent F/E testing. Two patients (0.24% of total, 2.3% of F/E group) sustained stable bony injuries detected only on F/E views, but all of these had other injuries detected on routine cervical imaging.
Message: While F/E films have been recommended for patients with persistent neck pain and normal radiographs, the actual contribution of these to decision-making is marginal at best.
Indications For CT? Computed tomography (CT) scanning has proven to be an excellent method for evaluating cervical spine fractures and dislocations. Its advantages include speed, wide availability, axial imaging, and excellent cortical detail. Contemporary high-resolution scanners detect between 95% and 100% of cervical fractures—a significantly higher sensitivity than plain-film radiography. So, when should this effective, albeit expensive, technology be used? The traditional approach would reserve CT imaging: 1) to delineate bony anatomy at the level of identified or suspected fractures and dislocations; 2) for those cases in which the upper or lower cervical spine cannot be adequately visualized; and 3) for patients with persistent pain and/or neurologic deficit despite normal plain films. A more aggressive strategy suggests that complete cervical helical scanning may be appropriate and cost-effective in severely injured patient who is at high-risk for cervical fracture.11 In the majority of cases a selective approach seems reasonable. Local practice should be driven collaboratively by emergency physicians, trauma and spine surgeons, and radiologists.
Special mention should be made of the intubated patient. Because the presence of an endotracheal tube may alter the radiographic appearance of upper cervical spine anatomy, a significant number of high cervical injuries may be missed on plain films. Several authors have suggested that patients undergoing cranial tomography for the evaluation of traumatic brain injury should have CT imaging extending through the upper cervical spine (C1 and C2).12,13
|Table: Summary Recommendations|
|•||Obtain cervical radiography in blunt trauma patients with:|
|—||Neck pain and midline cervical tenderness|
|—||Altered mental status, including intoxication|
|—||Focal neurological deficits|
|—||Distracting, painful injury|
|•||A 3-view series should be considered the "minimal standard"|
|•||Obtain oblique views when the 3-view is inconclusive, or the cervicothoracic junction is not well-visualized|
|•||Flexion-extension films are indicated in the patient with normal radiographs and suspected ligamentous injury|
|•||Use CT selectively:|
|—||To delineate anatomy at the level of injury|
|—||To define areas not well-visualized with plain films|
|—||With neurologic deficit / persistent pain and normal films|
|—||To assess C1-C2 in the intubated patient|
1. Hoffman JR, et al. Validity of a set of clinical criteria to rule out injury to the cervical spine in patients with blunt trauma. N Engl J Med 2000;343:94-99.
2. Mahadevan S, et al. Interrater reliability of cervical spine injury criteria in patients with blunt trauma. Ann Emerg Med 1998;31:197-201.
3. Ullrich A, et al. Distracting painful injuries associated with cervical spinal injuries in blunt trauma. Acad Emerg Med 2001;8:25-29.
4. MacDonald RL, et al. Diagnosis of cervical spine injury in motor vehicle crash victims: How may x-rays are enough? J Trauma 1990;30:392-397.
5. Turetsky DB, et al. Technique of use of supine oblique views in acute cervical spine trauma. Ann Emerg Med 1993;22:685-688.
6. Freemyer B, et al. Comparison of five-view and three-view cervical spine series in the evaluation of patients with cervical trauma. Ann Emerg Med 1989;18:818.
7. Nichols CG, et al. Evaluation of cervicothoracic junction injury. Ann Emerg Med 1987;16:640-642.
8. Davis JW, et al. Cervical injuries—perils of the swimmer’s view: Case report. J Trauma 1989;29:891-893.
9. Ireland AJ, et al. Do supine oblique views provide better imaging of the cervicothoracic junction than swimmer’s views? J Accid Emerg Med 1998;15:151-154.
10. Lewis LM, et al. Flexion-extension views in the evaluation of cervical-spine injuries. Ann Emerg Med 1991;20: 117-121.
11. Berne J, et al. Value of complete cervical helical computed tomographic scanning in identifying cervical spine injury in the unevaluable blunt trauma patient with multiple injuries: A prospective study. J Trauma 1999;47:896.
12. Link TM, et al. Substantial head trauma: Value of routine CT examination of the cervicocranium. Radiology 1995;196:741-745.
13. Blacksin MF, et al. Frequency and significance of fractures of the upper cervical spine detected by CT in patients with severe neck trauma. Am J Roentgenol 1995;165:1201-1204.