INTRODUCTION — Diagnosis of cervical spine injury in patients following trauma involves imaging. Cervical spine injuries can range from those that are minor and stable to more severe injuries that involve vertebral fractures or damage to the spinal cord, nerve root, ligaments, or vessels. This topic describes cervical spine imaging in adults including the choice of modality, image-acquisition procedures, and diagnostic performance of the imaging examinations. Related UpToDate content discusses:

●(See "Evaluation and initial management of cervical spinal column injuries in adults".)

●(See "Spinal column injuries in adults: Definitions, mechanisms, and radiographs".)

●(See "Acute traumatic spinal cord injury".)

INDICATIONS FOR IMAGING — Selection of patients to undergo imaging for suspected cervical spine trauma is determined by the mechanism of trauma and the patient's clinical signs and symptoms. Those considered at high risk for injury are imaged. For those considered at lower but not negligible risk, clinical decision rules (eg, National Emergency X-Radiography Utilization Study [NEXUS] criteria, Canadian C-spine Rule) are used when applicable to identify those who require imaging. Clinical evaluation to select patients for cervical spine imaging is discussed in greater detail separately, and an algorithm outlining our approach is provided (algorithm 1). (See "Evaluation and initial management of cervical spinal column injuries in adults", section on 'Determining the need for imaging'.)

CHOICE OF INITIAL IMAGING MODALITY

Computed tomography preferred — For patients who require imaging for suspected cervical spine injury, cervical spine computed tomography (CT) without contrast is the preferred imaging examination. The technology and expertise are available at most sites and around the clock at the major trauma centers. Cervical spine radiography is only used for the initial evaluation if the clinician is operating in a resource-limited setting without ready access to CT and without the capacity to transfer the patient to a facility with CT.

Cervical spine magnetic resonance imaging (MRI), computed tomography angiography (CTA), or magnetic resonance angiography (MRA) are not routinely performed in patients with suspected spine injury. They are reserved for further evaluation of some patients following the initial evaluation with CT.

Cervical spine CT is preferred over radiography because of its superior diagnostic performance and safety. A meta-analysis of 3832 patients imaged for suspected cervical spine trauma calculated the pooled sensitivity of radiography and CT as 52 percent (95% CI 47-56 percent) and 98 percent (95% CI 96-99 percent), respectively, for identifying those with injury [1]. Moreover, evaluating the spine with multidetector CT is faster, involves less patient movement, and results in far fewer technical failures requiring repeat imaging than with radiography.

The American College of Radiology (ACR) Appropriateness Criteria rate CT as "usually appropriate" and above radiography in patients who have undergone blunt cervical spine trauma and meet criteria for imaging [2]. Practice guidelines from the Eastern Association for the Surgery of Trauma (EAST) recommend that multidetector CT be used as the primary method of screening for cervical trauma [3].

CT performance and test characteristics — A cervical spine CT is performed on a multidetector scanner capable of volumetric acquisition. Patient is supine with the neck in a neutral (ie, neither flexion nor extension) position. Images are acquired with ≤3 mm collimation (typically ≤1 mm collimation) from the skull base through C7-T1 disc space. Thin-slice (≤1 mm) axial images in bone algorithm are preferred for primary evaluation in the trauma setting. Axial, sagittal, and coronal reconstructions are also performed in both bone and soft tissue algorithm. No intravenous contrast is given. Imaging in a nonhelical (ie, single-detector) scanner with axial acquisition is not sufficient to yield multiplanar (ie, sagittal and coronal) reformation images of sufficient quality for accurate interpretation.

For all patients who have sustained major trauma and are undergoing CT to assess for internal injury of the head, chest, or abdomen, cervical spine CT images are all acquired in the same sitting.

If a cervical spine fracture is detected, the entire spinal column (ie, cervical, thoracic, and lumbar spine), in general, is imaged. Fractures found at one level of the spine are associated with injury at other noncontiguous levels in approximately 20 percent of trauma patients [4,5]. This issue is discussed separately. (See "Evaluation of thoracic and lumbar spinal column injury", section on 'Decision rules for imaging thoracic or lumbar spine injury'.)

Several studies have reported that a multidetector CT with multiplanar reformations has a sensitivity and specificity approximating 100 percent for a clinically significant injury in an alert and cooperative patient [6-9]. As one example, a prospective multicenter trial performed by the Western Trauma Association involving 10,276 blunt trauma patients who did not meet National Emergency X-Radiography Utilization Study (NEXUS) exclusion criteria reported a sensitivity of 98.5 percent and a specificity of 91 percent, with a negative predictive value of 99.97 percent, for a clinically significant cervical spine injury (n = 198; defined as requiring surgery or stabilization with a halo or cervical-thoracic brace) [8]. Three false negative results were reported, all in patients with focal neurologic findings during their initial examination consistent with central cord syndrome. In another prospective study of 1668 intoxicated blunt trauma patients, researchers reported a negative predictive value of 99.8 percent for clinically significant cervical spine injury [10].

Although CT detects almost all fractures, it does not adequately image soft tissues such as ligaments, intervertebral discs, the spinal cord, and nerve roots. Thus, in patients with persistent or localizing signs and symptoms and a normal cervical spine CT, cervical spine MRI should be obtained. (See 'Further evaluation with magnetic resonance imaging' below.)

Estimated radiation dose from cervical spine CT is 5 to 6 mSv.

CT interpretation — Image interpretation requires evaluation of the axial, sagittal and coronal images reconstructed in bone and soft tissue algorithm. It is not sufficient to view images reconstructed in a soft tissue kernel in bone windows.

Assessment of bones in all three planes is required for accurate fracture diagnosis. Overall, spinal alignment is best assessed on the midsagittal view. For evaluating the specific articulations, the following planes are the most helpful:

●Lateral condyles of occipital bone, C1 and C2 lateral masses – Axial, parasagittal and coronal

●Intervertebral disc spaces – Sagittal and coronal

●Facet joints – Parasagittal and axial

●C2 dens – Coronal

Finally, the prevertebral soft tissues are evaluated in soft tissue windows on the midsagittal image. Soft tissue contour should parallel the vertebral bodies and be uniformly thin from C1 to C4. This enables indirect detection of underlying edema or hematoma.

Radiography — Cervical spine radiographs for trauma should include anterior-posterior (AP), lateral, and odontoid (ie, open-mouth) views. The lateral view should include the C7-T1 junction, which may require addition of a "swimmer's" view. On the lateral view, the entire cervical spine, from the base of the occiput to the top of the first thoracic vertebra, must be clearly visible. On the odontoid view, the dens and both lateral masses must be clearly visible (image 1).

While the lateral view is helpful in diagnosing spinal injuries, it is inadequate when used alone, detecting only 60 to 80 percent of fractures seen with a complete radiographic series [11,12]. Diagnostic yield increases substantially when AP and odontoid views are added.

In patients following major trauma or those who are obtunded for other reasons, adequate positioning to visualize the entire cervical spine is often not possible. Thus, radiography of the complete cervical spine with technically acceptable views cannot be obtained in up to 72 percent of patients [13].

Cervical spine radiographs identify only about one-third of fractures detected with CT [14]. If an adequate CT has been performed, radiographs are not needed for the initial evaluation, as they provide no additional information [15]. Patients with negative radiographs but persistent examination findings concerning for injury (eg, focal midline cervical tenderness, neurologic findings localizing to the upper extremities) should be managed with the presumptive diagnosis of a cervical spine injury.

Estimated radiation dose from cervical spine radiography is 1.5 to 2.0 mSv.

Radiography interpretation — Image interpretation begins with an assessment of whether adequate views have been obtained. If the images are adequate, the alignment, appearance, position, and spacing between vertebrae are evaluated. The soft tissues are assessed for evidence of underlying edema or hematoma. Guides for interpreting each of the three major views (lateral, odontoid, and AP) are provided (figure 1 and image 1 and image 2).

Loss of cervical lordosis alone is not sufficient to diagnose a cervical spine injury in the absence of the positive findings [16].

Flexion and extension radiography — Flexion and extension radiography in the trauma patient is not recommended. Optimum assessment and safe positioning requires a patient who is alert, cooperative, and without neck pain; however, adequate spinal movement is often not achievable due to pain and muscle spasm in the acute setting, and between 30 to 95 percent of flexion and extension radiographs are technically inadequate to exclude ligamentous injury [17-20].

FURTHER EVALUATION WITH MAGNETIC RESONANCE IMAGING — Cervical spine magnetic resonance imaging (MRI) without contrast is used to further evaluate patients with suspected ligamentous or spinal cord injury following computed tomography (CT). Lateral radiographs of the cervical spine with the neck in flexion and extension should NOT be performed in the acute setting. [1,2,11-13,15,21-26]

Indications for MRI — If CT suggests underlying ligamentous or spinal cord injury, our approach is to obtain MRI to evaluate the soft tissues, including the spinal cord, ligaments, disc, and nerve roots. Whether additional imaging is useful in patients with an isolated spinal cord injury after a negative CT remains controversial [2,7,27-29]. However, MRI can readily detect cord compression from discs or epidural hematomas that are difficult or impossible to identify with CT. MRI may detect ligamentous injury indicating spinal instability. In addition, MRI is used to assist with operative planning.

Suspected spinal cord injury without radiographic abnormality — Clinicians should suspect a cervical ligamentous or spinal cord injury without radiographic abnormality (SCIWORA) in an alert patient with severe neck pain, persistent midline tenderness, or upper extremity paresthesias; or if focal neurologic findings (eg, upper extremity weakness) are present despite a normal CT examination. (See "Evaluation and initial management of cervical spinal column injuries in adults", section on 'Suspected spinal cord injury without radiographic abnormality'.)

SCIWORA is often defined as the presence of neurologic deficits in the absence of evidence of bony injury on a complete, technically adequate radiography examination or CT. True SCIWORA is seldom associated with permanent neurologic injury in adults [30]; however, pure cervical spinal subluxation may occur when the ligamentous complexes rupture without an associated bony injury.

MRI performance and safety — For cervical spine MRI, the patient is supine with the neck in neutral (ie, neither flexion nor extension) position. The examination should be performed on a ≥1.5T scanner, and imaging should extend from the skull base to the C7-T1 interspace. While the imaging protocol varies somewhat with each practice and scanner, sagittal T1, T2-weighted fat-suppressed (eg, fast spin echo inversion recovery) sequences, axial T2-weighted sequences with or without fat suppression, and hemorrhage-sensitive (eg, gradient echo) sequences are obtained. Sagittal fat-suppressed T2-weighted or short TI inversion recovery (STIR) images are necessary in order to evaluate for bone marrow edema related to fracture and edema in the soft tissues due to ligamentous injury. The spinal cord is assessed for compression, edema, contusion, and hemorrhage on T1 and T2 sagittal as well as axial images with hemorrhagic products best seen on gradient echo images. Intravenous contrast is not required. Approximately 20 to 40 minutes are required for image acquisition.

MRI is more sensitive than CT for detecting soft tissue injury, but its specificity for detecting clinically significant discoligamentous injury remains low. The false-positive rate of MRI is estimated between 20 and 40 percent for detecting clinically significant injuries [31-33]. This may result in unnecessary morbidity. For example, in obtunded patients in the intensive care unit following trauma, adding MRI to CT for cervical spine clearance prolongs the period of rigid collar immobilization and mechanical ventilation, thereby increasing the risks for associated morbidities (eg, pressure ulcers, dysphagia, decreased cerebrovascular perfusion, elevated intracranial pressure, thromboembolism, ventilated associated infections, pressure ulcers, etc) [31].

If the patient is unresponsive, determining that he or she is free of indwelling electromagnetic or metallic implants to undergo MRI safely sometimes may not be feasible. In such cases, available CT and/or plain-film imaging as well as the family or health care proxy should be consulted to clear the patient. In the absence of health care history, minimally plain films of the skull, neck, chest, abdomen, and pelvis should be obtained to clear the patient before MRI. Obtunded patients should be monitored while in the scanner and ventilators, monitors, and other devices must be magnetic-resonance compatible [31,34,35]. (See "Patient evaluation for metallic or electrical implants, devices, or foreign bodies before magnetic resonance imaging".)

[17,19,24,36,37]

CTA OR MRA — Computed tomography angiography (CTA) or magnetic resonance angiography (MRA) is used to identify cerebrovascular injuries [2]. As the diagnostic performance is comparable, the choice is driven by availability of imaging resources and patient contraindications. CTA requires administration of intravenous contrast. MRA requires administration of gadolinium-based contrast for optimum accuracy but can be performed without contrast in patients with contraindications. Evaluation of trauma patients with suspected vascular injury is described elsewhere. (See "Blunt cerebrovascular injury: Mechanisms, screening, and diagnostic evaluation".)

In patients with trauma, CTA is generally preferred over MRA for logistical reasons. MRA involves longer table times, which, in an obtunded patient who cannot cooperate, will result in motion artifacts and poor image quality. Patient monitoring and support equipment are usually not magnetic-resonance compatible, which poses another hurdle. Thus, MRA is typically reserved for patients with contraindications to iodinated contrast or in patients who are undergoing another magnetic resonance exam for a separate indication (eg, brain or cervical spine magnetic resonance imaging [MRI] for injury).

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Upper spine and neck disorders" and "Society guideline links: Cervical spine injury".)

SUMMARY AND RECOMMENDATIONS

●Selection of patients to undergo imaging for suspected cervical spine trauma is determined by the mechanism of trauma and the patient's clinical signs and symptoms. Those considered at high risk for injury are imaged. For those considered at lower but not negligible risk, clinical decision rules (eg, National Emergency X-Radiography Utilization Study [NEXUS] criteria, Canadian C-spine Rule) are applied to identify those who require imaging. Clinical evaluation to select patients for cervical spine imaging is discussed separately, and an algorithm outlining our approach is provided (algorithm 1). (See "Evaluation and initial management of cervical spinal column injuries in adults", section on 'Determining the need for imaging'.)

●In patients for whom imaging is indicated, cervical spine computed tomography (CT) without contrast is the preferred examination. The technology and expertise are available at most sites and around the clock at the major trauma centers. Cervical spine radiography is only used if the clinician is operating in a resource-limited setting without ready access to CT and without the capacity to transfer the patient to a facility with CT. (See 'Choice of initial imaging modality' above.)

●Cervical spine magnetic resonance imaging (MRI) without contrast is used to further evaluate patients with suspected ligamentous or spinal cord injury following CT. Lateral radiographs of the cervical spine with the neck in flexion and extension should NOT be performed in the acute setting. (See 'Further evaluation with magnetic resonance imaging' above.)

●Whether additional imaging is useful in patients with suspected isolated ligamentous injury or spinal cord injury without radiographic abnormality (SCIWORA) after a negative CT is controversial. Our approach is to obtain cervical spine MRI without contrast to evaluate for ligamentous and disc injury as well as compressive hematomas in this setting. An alternative approach is to forego the MRI. (See 'Indications for MRI' above.)

●Computed tomography angiography (CTA) or magnetic resonance angiography (MRA) is used to identify cerebrovascular injuries. Choice is driven by availability of imaging resources and patient contraindications. Evaluation of trauma patients with suspected vascular injury is described elsewhere. (See "Blunt cerebrovascular injury: Mechanisms, screening, and diagnostic evaluation".)