Share this article on:

Spine Imaging: CT vs. Plain Films

Bukata, W. Richard MD

Journal Scan

Dr. Bukata is the medical director of the emergency department at San Gabriel Valley Medical Center in San Gabriel, CA, a clinical professor in the department of emergency medicine at Los Angeles County/USC Medical Center, and the editor of Emergency Medical Abstracts



Intuitively, it seems there should be no comparison of plain films with CT films, and as suspected, the literature points out that plain films miss a lot of fractures. There seems to be a disconnect, however, because the medical community still relies on plain films about 95 percent of the time (my estimate). Perhaps one of the reasons that the vast majority of neck imaging is with plain films is because we are really not suspicious that a fracture exists in 95 percent of the cases, but we take the films for the usual reasons: “The patient expects them,” “You never know,” “If I don't take them, the next doctor will,” and “It's the standard of care.”

When you really want to know the answer, however, there are enough papers that make it clear that the CT is what to order. It is very fast, doesn't require the shoulders to be pulled down to expose the lower vertebrae of the neck, allows soft tissue to be seen better, and is a much better study.

The following paper presents some compelling data in a retrospective look at 3,537 trauma patients, 236 with spinal fractures. The sensitivity of CT was 99%, and for plain films was 63%. Noncontinuous fractures also were common (19%).

Spiral Computed Tomography for the Diagnosis of Cervical, Thoracic, and Lumbar Spine Fractures: Its Time Has Come, Brown CVR, et al, J Trauma, 2005;58:890

BACKGROUND: There is a growing body of literature attesting to the superiority of spiral CT scanning over plain x-rays for cervical spine screening in blunt trauma patients, but there is only limited information on CT screening of the thoracic and lumbar spine.

METHODS: The authors reviewed data on spinal screening in 3,537 blunt trauma patients included in the Scripps Mercy Hospital (San Diego) Trauma Registry. Spiral CT scanning with sagittal and coronal reconstruction was performed in symptomatic patients and patients who could not be examined clinically. Patients with persistent neck or spine pain or tenderness underwent flexion-extension or MRI studies. Adjunctive plain x-rays were used selectively early after implementation of the CT screening protocol.

CT scans to assess spine fractures don't require the shoulders to be pulled down and allow soft tissue to be seen better

RESULTS: Spinal fractures were diagnosed in 236 patients (7%); fracture sites were the lumbar spine in 48 percent of these injuries, the cervical spine in 43 percent, and the thoracic spine in 28 percent (19% of the patients had injuries in more than one anatomical region). Spinal CT scanning failed to identify only two fractures (a C-spine compression fracture identified on MRI and treated with a cervical collar and a compression fracture of the thoracic spine identified on plain films requiring no treatment). The sensitivities of spiral CT scanning and plain x-rays (performed in 33 fracture patients) were 99% vs. 63%, respectively, for overall spinal fractures, 99% vs. 67% for C-spine fractures, 98.5% vs. 64% for thoracic spine fractures, and 100% vs. 69% for lumbar spine fractures.

CONCLUSIONS: These findings support the use of spiral CT scanning to screen for spinal fractures in blunt trauma patients, and suggest that routine plain film screening is no longer necessary.

Back to Top | Article Outline

Cervical Spine Injury

Here's another paper that makes it clear that CT is superior to plain films. It's a meta-analysis of seven studies involving 3,834 blunt trauma patients with cervical spine injuries. The pooled sensitivity for plain films was a miserable 52% vs. 98% for CT.

Computed Tomography versus Plain Radiography to Screen for Cervical Spine Injury, Holmes JF, et al, J Trauma, 2005;58:902

BACKGROUND: About four percent of injured patients admitted to U.S. trauma centers have a cervical spine injury. Although plain x-rays have been used traditionally to screen for C-spine injury, CT scanning is being used with increasing frequency.

CT scans show superiority to plain films for unconscious, intubated trauma patients

METHODS: The authors from UC Davis School of Medicine performed a meta-analysis of seven studies comparing plain radiography (at least an AP, lateral, or open-mouth odontoid view) and CT scanning (from the occiput to the superior aspect of the first thoracic vertebra) to screen for C-spine injury in 3,834 blunt trauma patients considered to require imaging.

RESULTS: None of the studies was a randomized, controlled trial. Five had minimal to moderate selection bias or lacked an independent gold standard, and two included fewer than 50 subjects or had severe selection bias. Several studies included only severely injured patients. The prevalence of a C-spine injury was five percent to 23 percent. In most of the studies, the gold standard included CT findings. The pooled sensitivity for C-spine injury was 52% for plain radiography (95% CI 47–56%) compared with 98% for CT scanning (95% CI 96–99%). Due to limitations of the available data, specificity, and positive and negative likelihood ratios could not be calculated.

CONCLUSIONS: The authors feel this limited evidence supports the use of CT scanning as the initial screening test for C-spine injury in patients at very high risk (e.g., those with significantly depressed mental status), but is insufficient to suggest that CT scanning should replace plain x-rays as an initial screening test in patients who are less severely injured.

Back to Top | Article Outline

CT in Trauma Patients

The third paper in this series also makes it clear that CT is substantially superior to plain films. It involved 437 unconscious, intubated trauma patients. CT was 98.1% sensitive while plain films were 72.1% sensitive. It would appear to me that any patient with a reasonable likelihood for a cervical spine fracture should get a CT (the miss rate on plain films seems unacceptably high) and that if a trauma patient is going to be getting a head CT, it would seem prudent and reasonable to get a cervical spine CT at the same time. The same would be true for a blunt trauma patient getting an abdominal CT: include the spine as well. Plain films seem to be great for patients who don't have spine injuries.

Helical Computed Tomographic Scanning for the Evaluation of the Cervical Spine in the Unconscious, Intubated Trauma Patient, Brohi K, et al, J Trauma, 2005;58(5):897

BACKGROUND: The incidence of unstable spinal injuries has been reported to be 4.6 percent to 34.4 percent in unconscious blunt trauma patients. The sensitivity of plain spinal x-rays in intubated patients is as low as 50 percent to 60 percent. Helical CT scanning does not directly visualize ligamentous injury, but is indirectly suggestive via identification of misalignment or malrotation of bony elements.

METHODS: This prospective British study reports on a protocol for spinal screening in 437 unconscious, intubated blunt trauma patients. Such patients first underwent a single cross-table lateral C-spine x-ray with maximal caudal traction on the arms (inadequate films were not repeated), followed by single-slice helical CT scanning from the base of the skull to T1 with sagittal and coronal reformation. MRI was performed selectively (e.g., if prior imaging was contradictory or suggestive of ligamentous injury or instability).

CT scans are superior to plain films for blunt trauma patients with cervical spine injuries

RESULTS: C-spine injuries were diagnosed in 14 percent of the patients, half of which (7% of the series) were unstable. CT scanning identified 60 of 61 C-spine injuries and all unstable injuries (the missed injury was visible on the plain film), yielding a sensitivity and negative predictive values of 98.1% and 99.7% overall, and 100% each for unstable C-spine injuries. Plain x-rays were adequate in 200 of 421 patients in whom they were performed; compared with CT scanning the overall sensitivity and negative predictive values were 72.1% and 95.2%, respectively. Using this protocol, discontinuation of spinal immobilization immediately after imaging was possible for 79.9 percent of the patients without spinal injuries.

CONCLUSIONS: The authors feel that helical CT scanning of the entire C-spine is currently the best screening modality for the unconscious trauma patient.

Back to Top | Article Outline

Assessing Tube Placement

Here's a prehospital care paper that makes an important observation: Patients intubated in the field should have objective assessment of tube placement. Certainly this would seem mandatory. If anesthesiologists are required to use end-tidal CO2 monitoring in the controlled setting of the operating room, certainly paramedics intubating in the worst conditions possible should be required to use devices to ensure proper tube placement. A tube in the wrong hole is likely to be fatal, a mistake that is easily avoided with current technology.

The Effectiveness of Out-of-Hospital Use of Continuous End-Tidal Carbon Dioxide Monitoring on the Rate of Unrecognized Misplaced Intubation Within a Regional Emergency Medical Services System, Silvestri S, et al, Ann Emerg Med, 2005;45(5):497

BACKGROUND: Some studies have reported rates of unrecognized misplaced endotracheal tubes in the prehospital setting ranging from about seven percent to 25 percent.

METHODS: This prospective observational study from the Orlando (FL) Regional Medical Center and the Orange County (FL) EMS (one author is a consultant for Oridion Medical, a capnography manufacturer) examined rates of misplaced endotracheal tubes upon arrival at a regional trauma center in 153 patients managed in the prehospital setting by one of 50 advanced life support emergency services agencies. Protocols for end-tidal CO2 (ETCO2) monitoring were not standardized, and capnography was performed at the discretion of the treating paramedics.

RESULTS: Of the total patient population, 70 percent were men (mean age, 43), 32 percent were medical patients, and 68 percent were trauma patients. Continuous ETCO2 monitoring was performed in 61 percent of the cases. The overall rate of misplaced endotracheal tubes upon ED arrival, according to the receiving physician, was nine percent overall, 23 percent in patients not managed with continuous ETCO2 monitoring, and zero in those managed with ETCO2 monitoring (odds ratio in patients without ETCO2 monitoring, 28.6). Eleven of 13 patients with unrecognized esophageal intubation died (9) or sustained severe neurologic impairment (2).

Patients intubated in the field should be objectively assessed for tube placement

CONCLUSIONS: Failure to implement continuous ETCO2 monitoring for patients receiving prehospital endotracheal intubation was associated with a high rate of misplaced ET tubes. This complication was not observed in any patient having continuous ETCO2 monitoring.

© 2006 Lippincott Williams & Wilkins, Inc.