CH 26

C H A P T E R zzzzzzzzzzzzzzzzzzzzzzzzzzz
26
Spinal Imaging Spinal Imaging
zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz
Stuart E. Mirvis, M.D., F.A.C.R.
Spinal imaging must be considered in the context of the clinical presentation of the entire patient, which dictates management priorities as well as the type and sequence of diagnostic imaging evaluations. For all acutely injured patients with clinical signs of spine or spinal cord injury and all noncommunicative patients in whom the account of the mechanism of injury is consistent with spine or spinal cord injury, at least a frontal (anteroposterior [AP]) and horizontal beam lateral radiographs of the spine must be obtained during the initial evaluation. If the potentially injured segment of the spine, especially the cervical spine, cannot be cleared (i.e., declared negative for injury) by this limited examination or if the patients condition requires immediate surgical intervention or more complex imaging procedures for other organ systems, the spine must be immobilized to protect the cord until the patient has been stabilized sufciently to complete denitive imaging examinations of the spine. Spinal imaging refers to evaluation of the spine by any of the various imaging modalities and techniques, or by any combination of such techniques, generally included in radiology. Diagnostic imaging of the spine is the denitive method for determining the presence, location, extent, and nature of injury to the spinal column, including, with the advent of magnetic resonance imaging (MRI), the spinal cord. Diagnostic imaging of the spine is therefore essential to the accurate assessment, evaluation, and management of spinal injury. The efcient and economic application of diagnostic imaging of spinal trauma requires a thorough knowledge of the indications for and limitations of the various imaging techniques available and the sequence in which they should be applied. The diagnostic imaging of each region of the spine is considered separately because of differences in anatomy and injury patterns. The rst section of this chapter considers currently available imaging modalities for evaluating acute cervical spinal injury along with illustrations of normal anatomy and examples of common injuries as they are related to the imaging concepts discussed. The following section describes approaches to the imaging evaluation of potential acute cervical spine trauma and
708
considers many of the controversies that surround this subject. This section begins with the initial plain lm study and includes the types and sequences of recommended imaging techniques, as dictated by the patients clinical condition, that culminate in a denitive diagnosis as soon as possible. The third section considers imaging of injury involving the noncervical spine. The fourth section reviews the strengths and weaknesses of each major imaging modality potentially used in assessing spinal injury. The nal section discusses imaging the traumatized spine of patients with preexisting conditions that may signicantly alter the normal appearance of the spine. The diagnostic approach presented reects experience in the imaging evaluation of acute spinal injuries within the environment of a Level I trauma center, but it is applicable to spinal injuries seen in any setting (Fig. 261).
DIAGNOSTIC IMAGING OF THE CERVICAL SPINE
zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz
Imaging Modalities: An Overview

The diagnostic imaging modalities useful in the evaluation of spinal trauma include plain lm radiography, computed tomography (CT), including two- and three-dimensional data re-formation, CT-myelography (CTM), MRI, and nuclear scintigraphy. Cervical vascular injury, with the potential for devastating neurologic consequences, is not uncommonly associated with cervical spine trauma. These potential vascular injuries may be recognized by CT using intravenous contrast or with routine MRI but are denitively diagnosed by magnetic resonance angiography (MRA), Doppler sonography, CT-angiography, and catheter angiography. The development and increasing availability of ultrafast multirow detector CT will make screening cervical CT-angiography practical for selected trauma patients at high risk for injury to the carotid and vertebral arteries.
Copyright 2003 Elsevier Science (USA). All rights reserved.
CHAPTER 26 Spinal Imaging
709
CLINICAL ASSESSMENT RESULT
Normal examination No cervical spine pain Alert No major distracting injuries
Neurologic deficit consistent with cervical cord/root injury
Neurologic examination normal Neck pain or tenderness to motion or palpation
Supine AP, OMO, lateral cervical radiographs1
Unreliable clinical assessment; obtunded, unconscious, major distracting injuries
Supine AP, OMO, lateral cervical radiographs2
Supine AP, lateral cervical radiographs3 Print Graphic
No imaging assessment recommended
Maintain cervical immobilization, obtain consultation to plan further evaluation
Abnormal
Normal
Normal Presentation
Physician supervised flexion-extension lateral radiographs
Total cervical spiral CT
Normal Normal Cervical MRI

1 Total
cervical CT may replace radiographic assessment in near future. 2 Must include C7 to top T1; swimmers lateral, supine obliques, or CT may be used. 3 OMO view typically difficult to obtain in this group.
Repeat flexion-extension views in 710 days if symptoms persist
Normal
Done
FIGURE 261. Algorithm for imaging diagnosis of cervical spine injury. AP, anteroposterior; CT, computed tomography; MRI, magnetic resonance imaging; OMO, open-mouth odontoid.
Plain Film Radiography of the Cervical Spine

All imaging examinations of spinal trauma currently begin with plain radiographic studies. Plain lm radiography is readily available in all emergency centers; it is a reliable and quick method for evaluation of patients and can be performed with portable or xed equipment. Radiography provides an excellent overview of the extent and magnitude of injury and makes a denitive and specic diagnosis possible in certain spinal injuries. The exibility of the x-ray tube-lm geometry provides the positioning latitude necessary for obtaining a comprehensive examination without motion of the patient, which is essential for patients in whom a spinal injury is suspected. The quality of the plain radiographic study is of paramount importance to the identication of cervical spine injury. A properly exposed radiograph must display both the skeleton and the soft tissues and must be free of motion or grid artifacts that could obscure or mimic
fractures. The prevention of artifact is of primary importance in detecting subtle, minimally displaced osseous injuries. A properly collimated plain lm study effectively limits patients exposure to radiation, and the examination is relatively inexpensive. LATERAL VIEW Evaluation of the initial screening lateral radiograph should be done methodically. An adequate study must include the cervicothoracic junction (to include the top of T1), and the patient should be positioned without rotation of the head. Every reasonable effort must be made to visualize the cervicothoracic junction on the initial plain radiographic examination. If the C7-T1 level is not adequately visualized on the lateral radiograph, the cervical spine cannot be cleared (i.e., declared negative), and other plain lm studies (see later discussion) or CT must be performed. The cross-table lateral radiograph is at least 74% to 86%
710
SECTION II Spine
sensitive for detection of cervical spine injuries, depending to a large extent on the expertise and experience of the examiner.33, 88, 125, 126 Missed injuries may result from (1) overlapping of bone, particularly involving the cervicocranial junction, the articular masses, and the laminae; (2) nondisplaced or minimally displaced fractures, particularly involving the atlas and axis; and (3) ligament injuries that may not be manifest when the radiograph is taken with the patient in a supine position with the neck in extension and stabilized by a cervical collar (i.e., no stress applied). Some cervical spine subluxations or dislocations can reduce spontaneously or be reduced with placement in a cervical collar before imaging evaluation, making their detection more difcult. Obviously, poor imaging technique related to positioning, exposure, or motion can signicantly impair diagnostic accuracy. Review of the lateral cervical radiograph involves assessment of anatomic lines, including the anterior and posterior vertebral margins, alignment of the articular masses, and alignment of the spinolaminar junctions (Fig. 262).57 It is important to recognize minimal degrees of anterior and posterior intervertebral subluxation that occur normally with physiologic motion with cervical exion and extension (Fig. 263). Such physiologic displacement typically occurs at multiple contiguous levels
Print Graphic
Presentation
FIGURE 262. Normal lateral cervical spine radiograph. Proper alignment of the cervical spine is seen as smooth continuity of the anterior vertebral margins (open black arrows), posterior vertebral margins (open white arrows), and spinolaminar junction lines (solid black arrows). The anterior atlantodental space (curved white arrow) should measure 2.5 to 3 mm or less.
and usually does not exceed 3 mm.132 The spacing between the laminae, articular facets, and spinous processes should be similar at contiguous levels (see Fig. 262). The intervertebral disc spaces should appear nearly uniform in height across the disc space. The orientation of each vertebra should be assessed for any rotational abnormalities. On a true lateral radiograph, the articular masses should be superimposed; abrupt offset of the masses indicates a rotational injury such as a unilateral facet dislocation. Similarly, an abrupt change in the distance from the posterior margin of the articular pillar to the spinolaminar line (the laminar space) also indicates a rotational injury158 (Fig. 264). Focal prevertebral or retropharyngeal soft tissue edema or hematoma can sometimes indicate an otherwise radiographically occult injury. However, absolute measurements of the prevertebral soft tissues are not particularly accurate indications of injury and can vary with head position, body habitus, and phase of inspiration, among other factors.65, 146 Herr and colleagues65 evaluated prevertebral soft tissue measurements at the C3 level in 212 patients with blunt trauma using a 4-mm upper limit of normal. They found that a measurement greater than 4 mm was only 64% sensitive for detecting cervical spine fractures involving the anterior, posterior, upper, or lower cervical spine. Precervical soft tissue prominence from the skull base to the axis is particularly important to recognize, as injuries at the craniocervical junction are often not apparent on the lateral radiograph. Harris56 has shown that the contour of the cervicocranial prevertebral soft tissues can be particularly useful in detecting subtle upper cervical spine injuries (Fig. 265). Assessment of the axis is aided by identication of the Harris ring,59 a composite shadow of cortical bone along the margins of the neurocentral synchondrosis (Fig. 266). In a true lateral projection, the Harris rings from both sides of C2 are superimposed, whereas two parallel Harris rings result from an oblique (off-lateral) projection. Also in the lateral projection, the ring of cortical bone is interrupted in its posteroinferior aspect by the smaller ring of the foramen transversarium. The Harris ring is particularly helpful in detecting atypical cases of traumatic spondylolisthesis8 (hangmans fracture) and the classical type III or low odontoid fracture2 (Fig. 267; see also Fig. 265). Radiologic identication of subluxation at the atlantooccipital articulation can be difcult. Previously, the Powers ratio60, 61, 117 was emphasized to assess this alignment. However, the anatomic landmarks required for this measurement are often difcult to visualize.60, 61 Alignment at this articulation can be assessed by reference to three anatomic landmarks in the neutral position. First, the occipital condyle should lie within the condylar fossae of the atlas ring, with no gap between them in the adult patient. Second, a line drawn along the posterior surface of the clivus should intercept the superior aspect of the odontoid process. Third, a line drawn along the C1 spinolaminar line should intercept the posterior margin of the foramen magnum (Fig. 268). A more precise assessment of this anatomic relationship can be determined by direct measurement, regardless of the degree of cervical exion and extension. The tip of the odontoid process should lie within 12 mm of the basion
711
FIGURE 263. Physiologic subluxation. Lateral cervical radiograph in exion (A) shows slight anterolisthesis at several levels (arrows). Relative lack of articular process overlap at C5C6 is also physiologic. Extension (B) view shows physiologic retrolisthesis at C2C3 and C3C4 (arrows).
Print Graphic
Presentation
(inferior tip of the clivus), and a vertical line drawn along the posterior cortex of C2 (posterior axial line) should lie within 12 mm of the basion60, 61 (Fig. 269). Cranial distraction and anterior displacement are indicated by measurements greater than 12 mm. The C1C2 articulation should be evaluated for the anterior atlantodental interval, normally less than 3 mm in the adult.21 A larger atlantodental interval associated with a cervicocranial hematoma indicates acute transverse atlantal ligament injury (Fig. 2610) and instability of the articulation. ANTEROPOSTERIOR VIEW The AP radiograph of the cervical spine supplements information provided on the lateral cervical radiograph and can identify additional injuries.32, 57, 125, 151 On the normal AP view, the spinous processes are vertically aligned, the lateral masses form smoothly undulating margins without abrupt interruption, the disc spaces are uniform in height from anterior to posterior, and the alignment of the vertebral bodies is easily assessed (Fig. 2611). Typically, the craniocervical junction region and the odontoid process are not visible, being obscured by the face, mandible, and occipital skull. Lateral translation (displacement) of the vertebral bodies is best appreciated in this view. Similarly, lateral exion injuries compressing a lateral mass or the lateral portion of a vertebral body are also demonstrated to advantage. Rotational injuries are indicated by an abrupt offset of spinous process alignment, as occurs with unilateral facet dislocation.
Fractures of the vertebral body in the sagittal plane are often evident on the AP view. Facet and articular mass fractures can sometimes be visualized as well. Laminopedicular separation (fracture separation of the articular mass), which may occur with hyperexion-rotation injuries134 or rarely hyperextension62 mechanisms, can produce a horizontal orientation of the articular mass, leading to an open-appearing facet joint (Fig. 2612). Normally, these joints are not seen in tangent on the AP view because of their 35-degree inclination from the horizontal plane. Some studies suggest that the AP view provides no signicant diagnostic information in addition to that available from the lateral and open-mouth projection.67 West and co-workers155 have shown that a single lateral cervical spine radiograph is as sensitive for injury diagnosis as the standard three-view series for experienced interpreters. OPEN-MOUTH ODONTOID VIEW The open-mouth odontoid (OMO) or atlantoaxial view requires cooperation on the part of the patient for optimal studies. Ideally, the skull base (occiput), atlas, and axis are well displayed without overlap from the mandible or dentition (Fig. 2613). The normal OMO view demonstrates the lateral margins of the C1 ring aligned within 1 or 2 mm of the articular masses of the axis. The articular masses of C2 should appear symmetric, as should the joint spaces between the articular masses of C1 and C2 as long as there is no rotation of the head. The measured distance between the odontoid and the C1 medial border (i.e., the
712
SECTION II Spine
lateral atlantodental space) should be equal, but a discrepancy of 3 mm or greater is often seen for patients without pathology.69 Finally, a vertical line bisecting the odontoid process should form a 90-degree angle with a line placed horizontally across the superior aspect of the C2 articular masses147 (Fig. 2614). Voluntary rotation, head tilting, or torticollis can be difcult to distinguish from atlantoaxial rotatory subluxation on the basis of radiography alone. Dynamic CT studies can be useful in differentiating a locked atlantoaxial dislocation from subluxation without locking.105 Injuries that are best seen on the OMO view include the C1 burst fracture (Jefferson fracture) (Fig. 2615), odontoid fractures (see Fig. 2614), and lateral exion fractures of the axis. Lateral spreading of the C1 lateral masses of greater than 6 to 7 mm in the Jefferson burst fracture suggests coexisting disruption of the transverse portion of the cruciate ligament, producing an unstable atypical Jefferson fracture.42 This pattern usually creates two fractures on one side of the C1 ring and probably results from asymmetric axial loading or bending forces (Fig. 2616).
SUPINE OBLIQUE (TRAUMA OBLIQUE) VIEW The supine oblique, or trauma oblique, projection is obtained with the patient maintained in collar stabilization in the supine and neutral position. The lm-screen cassette is placed next to the patients neck, and the x-ray tube is angled 45 degrees from the vertical.57 The normal oblique view shows the neural foramina on one side and the pedicles of the contralateral side. The laminae are normally aligned like shingles on a roof (Fig. 2617). This projection can be used to improve visualization of the cervicothoracic junction when the lateral view is insufcient and is often utilized to clear the cervicothoracic junction.72 Subluxation or dislocation of the articular masses and laminae that may not be seen on other standard views may be shown to advantage in this projection. If the cervicothoracic junction cannot be adequately visualized on the neutral lateral view, most institutions obtain a swimmers lateral radiograph (89%) as opposed to bilateral supine oblique (11%) as the next imaging study.71 The supine oblique views are costeffective compared with CT scanning for selective clearing of the cervicothoracic junction.72 SWIMMERS LATERAL PROJECTION This view is often acquired to visualize the cervicothoracic junction when it is obscured by the density of shadows produced by the shoulders in the true lateral projection. Optimal positioning requires that one of the patients arms be abducted 180 degrees and extended above the head which may be difcult or impossible in patients with arm and shoulder injurieswhile the opposite shoulder is extended posteriorly to decrease overlapping of skeletal structures (Fig. 2618).57 The projection further requires that the patient be rotated slightly off the true lateral. Positional changes required to obtain the swimmers view are contraindicated for patients who are unconscious or who have cervical cord injuries. The swimmers view results in a somewhat distorted oblique projection of the cervicocranial junction, with the vertebrae obscured by portions of the shoulder girdle or the ribs, or both. Even with its limitations, however, this view is generally suitable to assess alignment and detect gross injuries. A modication of the swimmers projection12 is designed to improve the quality of the image by producing a truer lateral projection. PILLAR VIEW The pillar view (Fig. 2619) is specically designed to visualize the cervical articular masses directly in the frontal projection. Weir152 contended that the pillar view should be included in all acute cervical injuries. It is generally agreed, however, that this view should be reserved for neurologically intact patients in whom articular mass fractures are suspected on the basis of radiography. The pillar view is obtained by rotating the patients head in one direction, off-centering the x-ray tube approximately 2 cm from the midline in the opposite direction, and angling the central x-ray beam approximately 30 degrees caudad, centered at the level of the superior
Print Graphic
Presentation
FIGURE 264. Unilateral facet dislocation. Lateral cervical radiograph shows anterior subluxation of C5 on C6. There is offset of the articular masses at C5 (single-headed arrows) and superimposition of these at C6, indicating rotation of the articular masses. There is an abrupt alteration in the distance from the back of the articular mass to the spinolaminar line (laminar space), also indicating rotation of the C5 vertebral body relative to C6 (double-headed arrows).
713
Print Graphic
Presentation
FIGURE 265. Subtle soft tissue abnormality indicating fracture. A, Coned-down lateral view of the craniocervical junction shows normal prevertebral soft tissue conguration with a slight soft tissue fullness (convex bulge) at the level of the C1 anterior arch and a slight concavity below (arrows). B, Coned-down lateral view from another patient with cervical pain shows loss of these contours with uniform prevertebral soft tissue fullness above, at, and below the C1 anterior tubercle. A subtle odontoid fracture (arrows) is present. The fracture interrupts the Harris composite ring shadow. C, Lateral cervical radiograph from another patient shows prevertebral soft tissue prominence at the cervicocranial junction with loss of normal contours around the C1 anterior tubercle (arrowheads), indicating prevertebral hematoma and edema. The spinolaminar line of C2 is also posteriorly displaced (solid arrows and black line). There is a subtle traumatic spondylolisthesis (open arrow). (B, From Mirvis, S.E.; Young, J.W.R. In: Mirvis, S.E.; Young, J.W.R., eds. Imaging in Trauma and Critical Care. Baltimore, Williams & Wilkins, 1992, p. 343.)
Print Graphic
Print Graphic
Presentation
Presentation
FIGURE 266. C2 composite ring shadow. In the lateral projection, portions of the C2 cortex form a composite shadow of bone density (arrows). A small circle of the foramen transversarium interrupts the ring in its posterior inferior margin (arrowhead). Discontinuity or irregularity of the ring shadow indicates probable type III odontoid fracture or atypical traumatic spondylolisthesis.
FIGURE 267. Type III odontoid fracture. Coned-down lateral view shows complete disruption of the Harris composite ring shadow, indicating odontoid fractures (arrows). The atlas is anteriorly displaced relative to the axis, as indicated by respective spinolaminar lines (arrowheads). (From Mirvis, S.E.; Shanmuganathan, K. J Intensive Care Med 10:15, 1995.)
714
SECTION II Spine
Print Graphic Print Graphic
Presentation
Presentation
FIGURE 268. Normal atlanto-occipital anatomic relationships. In a coned-down lateral view from a normal patient, the spinolaminar line of C1 aligns with the posterior foramen magnum (double-headed black arrow), and the occipital condyle sits within the fossae of the C1 ring without a gap (black arrowheads). The distance between the basion (tip of clivus, asterisk) and the top of the odontoid process (plus sign) is less than 12 mm. The posterior axial line (white arrows) drawn along the posterior margin of the C2 body lies within 12 mm of the basion.
FIGURE 2610. Atlantoaxial dislocation. Lateral coned-down cervical view shows marked widening of the atlantodental space, indicating disruption of the transverse atlantal ligament. Note the anterior displacement of the C1 spinolaminar junction line relative to that of C2. (From Mirvis, S.E.; Shanmuganathan, K. J Intensive Care Med 10:15, 1995.)
Print Graphic
Presentation
margin of the thyroid cartilage.57 The caudally angled central beam is tangential to the plane of the facet joints only in the middle and lower cervical spine because of normal cervical lordosis. Rotation of the head is essential to eliminate superimposition of the mandible on the lateral masses. Therefore, the patient must be able to rotate the head on command, and the presence of an upper cervical injury must have been previously excluded in the initial plain radiographic evaluation. Because articular pillar and pedicle fractures often occur with rotational injuries,135 further rotation is contraindicated when assessing these injuries. If injuries to the lateral cervical pillars are suspected on the basis of the initial plain lm screening, they are best assessed further by CT.
FLEXION-EXTENSION STRESS VIEWS

FIGURE 269. Atlanto-occipital subluxation. Lateral view of the upper cervical spine shows anterior displacement of the posterior margin of the foramen magnum relative to the C1 spinolaminar line (short arrows), the occipital condyles are displaced from the condylar fossae of the atlas (arrowheads), and a line drawn along the clival posterior margin intercepts the odontoid process along its anterior surface (long arrow). Finally, the distance between the tip of the odontoid (plus sign) and the basion (asterisk) exceeds 12 mm, as does the distance from the posterior axial line (interrupted line) to the basion.
Demonstration of ligament injury may require placing stress on the cervical ligaments. It is imperative that cervical exion and extension views be obtained only for alert, cooperative, and neurologically intact patients who can describe pain or early onset of any subjective neurologic symptoms. During the evaluation of acute injuries, active exion-extension radiographs should be supervised by a physician. The use of uoroscopically
715
Print Graphic
Print Graphic
Presentation
Presentation
FIGURE 2611. Normal anteroposterior (AP) cervical radiograph. In the normal AP view, a smoothly undulating lateral border is created by the lateral masses, the spinous processes are vertically aligned, and the spacing of the intervertebral discs and uncovertebral joints (arrows) is uniform. The facet joints cannot be visualized because they are inclined about 35 degrees to the horizontal. Enlarged C7 transverse processes (cervical ribs) are seen as a variant. In general, C2 and C1 are poorly seen in this projection. (From Mirvis, S.E.; Young, J.W.R. In: Mirvis, S.E.; Young, J.W.R., eds. Imaging in Trauma and Critical Care. Baltimore, Williams & Wilkins, 1992, p. 298.)
FIGURE 2612. Isolation of articular pillar. This anteroposterior view shows direct visualization of the facet articulations (arrowheads) of C5C6 and C6C7 on the left caused by rotation of the articular mass (arrow). This nding requires combined fractures of the lamina and ipsilateral pedicle.
guided passive exion-extension cervical spine assessment is discussed in detail in the following section on imaging approaches. Although for most patients evidence of cervical instability is apparent on the neutral lateral radiograph, some injuries can be effectively reduced to an anatomic position with the patient in a cervical collar and may be completely invisible in the stabilized neutral position. Normally, exion and extension produce minimal physiologic motion of adjacent vertebrae and anterior or posterior sliding movement across the articular facets. An abrupt change in facet coverage at one level indicates injury to the ligament support (Fig. 2620). Finally, degenerative disease (cervical spondylosis) of the facet articulations with loss of the interarticular brocartilage may allow excessive anterior translation at one or more levels that can mimic pathologic movement related to acute injury (Fig. 2621). In degenerative slippage, the shape of the articular facet and width of the facet joint spaces may be normal; however, in most cases the articular facet has become ground down, the facet joints are narrowed, and the articular processes are thinned. In
Print Graphic
Presentation
FIGURE 2613. Normal open-mouth odontoid (OMO) view. The lateral borders of the lateral masses of the atlas are aligned with the lateral borders of the axis. Without rotation of the head, the lateral atlantodental spaces are equivalent. A thin radiolucent Mach line at the base of the odontoid is created by the overlapping inferior surface of the posterior atlas ring (arrow).
716
SECTION II Spine
Print Graphic
Print Graphic
Presentation
Presentation
FIGURE 2614. Odontoid fracture with tilt. The open-mouth odontoid view shows a fracture across the base of the odontoid (arrows) and lateral tilting of the odontoid process.
FIGURE 2616. Unstable Jefferson fracture. Axial computed tomography (CT) scan through the atlas shows wide displacement of parts of the ring (double-headed arrows), indicating probable disruption or avulsion of the transverse atlantal ligament, creating both mechanical and neurologic instability.
traumatic subluxation, the articular facets are either normally shaped or fractured and the joint spaces are widened.82 In a national survey of 165 trauma centers, Grossman and associates51 found that exion-extension views were more likely to be obtained as part of the cervical spine imaging evaluation in Level I as opposed to Level II or lower level centers. Brady and colleagues15 evaluated use of dynamic exion-extension views in 451 patients with blunt trauma who manifested neck pain, midline tenderness, or an abnormal spinal contour on static
cervical radiographs. Patients with abnormal cervical static radiographs were statistically more likely to have abnormal active exion-extension studies than those with normal static studies and more likely to require invasive xation.
Print Graphic
Print Graphic
Presentation
Presentation
FIGURE 2615. Jefferson burst fracture. The open-mouth odontoid view shows lateral displacement of the C1 articular masses relative to those of C2 (arrows), indicating a C1 burst fracture.
FIGURE 2617. Normal shingles-on-the-roof orientation of laminae seen in an oblique projection of the cervical spine.
717
Print Graphic
Print Graphic
Presentation
Presentation
FIGURE 2619. Pillar view of the right articular masses of the cervical spine. FIGURE 2618. Swimmers view of the cervicothoracic junction.
IMAGING APPROACH TO THE POTENTIALLY INJURED CERVICAL SPINE
Spinal Imaging of the Polytraumatized Patient: An Overview

Acute injuries of the cervical spine range in signicance from clinically trivial to permanent paralysis. Therefore, the evaluation of cervical spinal injury must begin with clinical examination of the patient by an experienced physician who can appraise the location, extent, and magnitude of the spinal injury. Unconscious, intoxicated, or elderly patients with a history of trauma must be considered to have a cervical spinal injury until proved otherwise. The type and extent of imaging evaluation are governed by the patients neurologic status and overall condition, particularly the hemodynamic status. For all patients with spinal injury, however, the following guidelines should be observed: 1. The spine must remain immobilized to protect it and the cord during the initial examination and until the spine has been declared negative. 2. The radiologic evaluation begins with plain radiographs.
FIGURE 2620. Flexion subluxation. Flexion lateral radiograph shows a focal decrease in coverage of the articular facets at C5C6 (arrow). There is slight aring of the spinous processes at this level and minimal narrowing of the disc space compared with other levels. All ndings indicate hyperexion subluxation injury.
Print Graphic
Presentation
718
SECTION II Spine
Print Graphic
Presentation
frontal and horizontal beam lateral plain radiographs obtained in the emergency center, followed by CT with appropriate image re-formations and MRI of the injured area whenever clinically appropriate. Trauma patients with denite myelopathy but negative plain radiographs are best served by performing MRI or, if MRI is unavailable, myelography and CTM as the next study. Occasionally, the denitive evaluation of the spinal injury in major trauma patients must be superseded by the evaluation and management of more urgent clinical problems, such as a low Glasgow Coma Scale score, a tension pneumothorax, a suspected aorta injury, or massive or progressive hemoperitoneum. In such instances, and whenever possible, AP and lateral radiographs of the injured region of the spine should be obtained concurrently with initial evaluation of the patient and stabilization procedures. If the spine cannot be declared negative on the basis of this screening examination, it must be immobilized until the patient has been stabilized and the evaluation of the spinal injury can be concluded. If plain radiographs do not reveal the cause of the neurologic decit, then whenever possible, MRI should be obtained as the next examination. ALERT TRAUMA PATIENT WITH CERVICAL SPINE PAIN Radiographic examination of the spine is clearly not indicated for every patient who complains of minimal symptoms after minor trauma. The attending physician, however, must have a high index of suspicion regarding the presence of spinal injury, as failure to recognize and treat a clinically subtle dislocation, fracture, or fracturedislocation can lead to devastating, irreparable cord injury. Therefore, a history of trauma that could produce spinal injury or the presence of objective physical signs consistent with spinal injury is of particular importance in determining which patients should undergo imaging evaluation. The appropriate radiologic evaluation of the cervical spine is controversial, particularly with regard to the number of views that constitute an adequate radiologic assessment.* Whereas a single lateral radiograph detects 74% to 86% of cervical spine injuries, sensitivity increases to nearly 100% when an AP and an OMO projection are added.33 For the neurologically intact alert trauma patient with a complaint of neck pain, physical ndings of pain, or point tenderness elicited on palpation of the cervical spine, a minimum of two views, typically a cross-table lateral and an AP view, are obtained. Additional views, including the OMO projection and supine oblique views, can also be obtained and may increase diagnostic yield slightly. If the AP and lateral views obtained with the patient in a rigid collar are negative from the occiput to the C7-T1 level, the collar can be removed to facilitate obtaining the OMO and oblique views. If all plain radiographs are normal, the possibility of a signicant missed skeletal cervical spine injury is minimal. However, ligamentous injuries, including potentially neurologically unstable injuries such as reduced bifacet dislocations or reduced hyperexion or hyperextension
*See references 33, 62, 63, 68, 97, 125, 126, 133, 142, 150.
FIGURE 2621. Degenerative subluxation. Lateral cervical radiograph shows slight anterolisthesis of C4 on C5 (arrow). There is diffuse degenerative change. Note narrowing of the facet joints and sclerosis of the facet surfaces (arrowheads).
3. The initial examination must be monitored by a radiologist or other qualied physician to establish the radiologic diagnosis as accurately and quickly as possible and to determine whether additional examinations are indicated. If so, the most efcient sequence of studies for optimal care of the patient must be established. In the absence of neurologic ndings and with the spine immobilized, other injuries, if any, can be managed initially. Alternatively, it may be clinically appropriate to clarify the status of the spine before managing a coexistent injury. In any case, the examination of the spine must be personally supervised and sequentially monitored by a radiologist or other qualied physician until either the spine studies have been declared negative or a denitive diagnosis has been established.
Recommended Imaging Approach for Potential Cervical Spine Trauma Based on Clinical Presentation (see Fig. 261)
TRAUMA PATIENT WITH A NEUROLOGIC DEFICIT REFERABLE TO THE CERVICAL SPINAL CORD The optimal radiologic approach to the trauma patient with spinal injury and signs of cord damage consists of only
719
subluxations, may still not be demonstrated (Figs. 2622 and 2623). It is important to remember that on spinal radiographs obtained only with the patient supine, the patients body habitus may obscure some spinal injuries. Initial radiographs obtained with spinal immobilization in place should be examined by a physician skilled in the interpretation of spinal radiographs. At the University of Maryland Shock Trauma Center, clinically stable, alert, cooperative patients with cervical pain and normal initial cervical spine radiographs undergo exion-extension lateral radiography after the cervical collar is removed in order to assess ligamentous stability. On occasion, in the experience of that center, exion and extension subluxations may be identied that are not evident on neutral cross-table lateral or AP cervical radiographs obtained in the supine position.85 The use of routine exion and extension lateral cervical radiographs is by no means universal and should not be considered standard practice.51 This procedure should be carefully supervised by a physician and never performed by a radiography technologist alone. The patient should ex and extend the neck to the limit of pain tolerance or onset of subjective neurologic symptoms. Obviously, any suggestion of an onset of neurologic impairment mandates return to the neutral position and reapplication of cervical immobilization. If adequate exion and extension views are acquired with visualization of the spine through C7-T1, the vast majority of potentially unstable injuries are excluded. Flexion and extension lateral cervical views should not be obtained in uncooperative patients or those with decreased mental acuity and should not be obtained
with passive movement of the patients cervical spine by a physician. Alternatively, spiral CT can be used to screen patients with neck pain for subtle fractures that may be difcult or impossible to diagnose from radiographs. Increasingly, spiral CT is being used as a primary screening test in blunt trauma patients with cervical spine symptoms. Patients with neck pain who undergo CT scanning of other body regions may also undergo cervical spine CT to assess the region of neck pain or the entire cervical spine.106 Although it is well established that CT is more sensitive than plain radiography in detecting cervical spine injury,* a positive cost-benet ratio has not been determined for CT performed for radiographically negative trauma patients with neck pain. Increasingly, spiral CT and multidetector CT are being used to perform screening studies of the entire cervical spine rather than depending on plain radiographic interpretation for alert patients with cervical spine symptoms. This development is based on the recognized improved accuracy of CT over radiography for osseous pathology, the increased use of CT in general for assessing stable patients with blunt trauma in multiple body regions, the increasing speed of image acquisition and processing, and its general costefcacy.10, 13, 24, 55, 74, 83, 106, 107 The accuracy of spiral CT in detecting all potentially unstable ligament injuries is not known, although some of these injuries would be suggested by soft tissue swelling or subtle abnormalities of alignment. For this reason, active
*See references 1, 4, 16, 22, 35, 54, 55, 74, 87, 92, 106, 107, 116.
Print Graphic
Presentation
FIGURE 2622. Hyperexion ligament injury not apparent on neutral position lateral view radiograph. A, Lateral cervical radiograph in a trauma patient with cervical spine pain is unremarkable. B, Flexion lateral view (physician supervised) shows hyperexion subluxation at C5C6 (white arrow) and unilateral facet dislocation at C4C5 (black arrow). (A, B, From Mirvis, S.E.; Shanmuganathan, K. J Intensive Care Med 10:15, 1995.)
720
SECTION II Spine
Print Graphic
Presentation
FIGURE 2623. Hyperexion ligament injury not apparent on neutral position lateral cervical radiograph. A, Neutral position lateral radiograph of a trauma patient with cervical tenderness is unremarkable. B, Repeated view with exion limited by pain shows hyperexion sprain at C4C5 (arrowhead) with mild aring of the spinous processes (open arrow).
exion-extension views are still needed to ensure cervical spine ligament integrity. TRAUMA PATIENT WITH AN UNRELIABLE PHYSICAL EXAMINATION Trauma patients presenting without evidence of myelopathy but whose physical examination cannot be considered reliable constitute a major challenge with regard to possible spinal injury. All such patients should be regarded as potentially having unstable spinal injuries until proved otherwise. The radiographic assessment of such patients should include at least lateral and AP cervical spine views. Open-mouth views are often difcult to obtain and suboptimal technically in this population. In addition, supine oblique views of the cervicothoracic junction region can be obtained if needed. Often, CT is used to assess the cervicocranial junction if not well demonstrated by radiography, depending on its availability and indication for CT of other body regions. If injuries are identied, spine immobilization is maintained and further imaging workup performed when clinically feasible. The potential role of spiral CT scanning for screening the entire cervical spine for patients with an unreliable clinical examination is currently undergoing study. If all radiographic evaluations of the spine are negative, the vast majority of injuries are excluded, but again, the potential for a neurologically unstable injury persists. DAlise and colleagues26 performed limited cervical spine MRI within 48 hours of trauma in 121 patients who had no obvious injury shown by plain radiography. There were
31 patients (25.6%) who had signicant injury to paravertebral ligamentous structures, the intervertebral disc, or bone. Eight of these patients required surgical xation of the injury. If MRI is not available, an alternative approach is to obtain an erect AP and lateral cervical radiograph with the patient in collar stabilization to allow limited physiologic stress on the cervical spine. If these lms are normal, erect AP and lateral views are repeated out of collar with the cervical spine slightly extended and the head supported by a pillow. If these views are normal, the cervical collar is permanently removed. Although these approaches are considered prudent to avoid missing a cervical spine injury, they are by no means universally followed. In some sites the cervical collar is removed on the basis of negative AP and lateral supine radiography alone, suggesting the rarity with which neurologically unstable cervical spine injuries occur with normal-appearing cervical radiographs.52 Some authors have suggested the use of passive exion and extension imaging under uoroscopic guidance for patients who cannot have a reliable physical examination.27, 128, 131 Thus far, no complications of this procedure have been reported. However, almost all the patients studied have been normal, as would be expected from the extremely low pretest probability of an unstable injury. The limited data available do not provide sufcient evidence to support routine use of this technique. A number of cervical spine injuries are not detected by this method, including herniated intervertebral discs and epidural hematomas. These lesions may cause spinal cord compression that
721
creates or worsens a neurologic decit without evidence of overt subluxation on uoroscopy. Cervical disc herniation is a more common cause of central cord syndrome than previously suspected.25 Benzel and co-workers9 found 27 acute cervical disc herniations among 174 trauma patients with negative cervical radiographs who underwent cervical MRI. Rizzolo and colleagues124 observed acute cervical disc herniation in 42% of 55 patients with blunt cervical trauma with cervical fractures or neurologic decits. In addition, either congenital or acquired spinal stenosis can produce spinal cord lesions in association with blunt trauma when there is no radiographic evidence of injury.80 Flexion and extension in this population could worsen cord compression and ischemia. To date, only MRI has proven diagnostic accuracy for direct diagnosis of ligament injury from blunt spinal trauma.9, 26, 76, 79, 102, 110 ALERT TRAUMA PATIENT WITH NORMAL CERVICAL SPINE EXAMINATION The need to perform imaging in alert, appropriately oriented trauma victims without evidence of cervical pain, tenderness to palpation of the cervical spine, or major distracting injuries has been highly controversial. Most patients admitted to emergency centers from the scene of a major blunt force trauma are placed in cervical immobilization and are presumed to have a cervical injury until proved otherwise. This scenario places a great deal of pressure on the admitting physician to exclude an injury with an extremely high degree of certainty. There are case reports describing so-called painless cervical spine fractures. A close review of many such articles typically reveals that the patient either had symptoms or was not truly alert.33, 40, 68, 89, 95, 127 Many large series published to date indicate that alert trauma patients without major distracting injuries and without subjective complaints of neck pain or positive physical ndings invariably have normal imaging evaluations.33, 49, 67, 159 A prospective series of alert trauma patients without symptoms who underwent cervical spine CT to clear the cervicothoracic junction revealed one nondisplaced C7 transverse process fracture in 146 patients at a cost of more than $58,000.97 Diliberti and Lindsey33 recommended omission of radiologic assessment of the cervical spine in any trauma patient with class 1 level of consciousness (i.e., able to follow complex commands, responds immediately) and without evidence of intoxication, neurologic decit, cervical spine pain, or pain elicited on palpation. Gonzales and associates48 found that clinical assessment was more sensitive than radiography in detecting cervical spine injury even in intoxicated patients. CONCOMITANT CERVICAL SPINE AND LIFETHREATENING INJURIES As stated throughout this chapter, the imaging evaluation of the spine must be performed in the total context of the trauma patients management. The radiographic examination of patients with concomitant acute spinal injury and life-threatening injuries should consist of only AP and
horizontal beam lateral projections obtained in the emergency center during the clinical evaluation. If a radiologic diagnosis can be made from this limited study (e.g., traumatic spondylolisthesis, bilateral facet dislocation, burst fracture), management of the injury consistent with the patients clinical condition can be initiated. If results of the initial limited examination are equivocal or if the spinal injury is one that requires additional evaluation by CT or MRI, the spine must be appropriately immobilized until the life-threatening injury has been stabilized and the radiologic evaluation can continue.139
IMAGING ASSESSMENT OF POTENTIAL INJURY TO THE NONCERVICAL SPINE
zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz Alert trauma patients with pain in the thoracic, lumbar, or sacral region require lateral and AP radiologic views of the region in question. If these studies are negative but clinical symptoms are impressive, further imaging by CT is indicated. After radiologic identication of a thoracic, lumbar, or sacral fracture, CT is helpful in characterizing complex injuries such as fracture-dislocations and in distinguishing burst fractures from anterior compression fractures.58 Acute onset of radicular symptoms after acute trauma may also warrant CTM or MRI to exclude acute intervertebral disc herniation. In the patients who are not reliable enough for an accurate physical examination, AP and lateral spine lms also provide routine screening. The improvements in CT technology, introduced with spiral CT and the newer multidetector array systems, create the potential for CT to provide screening of the thoracic and lumbar spine as part of a routine thoracic cavity and abdominal-pelvic CT study in a multipletrauma patient. Single-slice or multislice spiral CT used in conjunction with scout AP and lateral radiographs of the spine may ultimately provide more accurate identication of thoracic and lumbosacral injuries than is achieved with conventional radiography.
Plain Film Radiography of the Thoracic Spine

Imaging of the thoracic spine is a less complex procedure than that of the cervical spine. The inuence of the patients overall condition on the type and sequence of imaging procedures used to evaluate the thoracic and lumbar areas is identical to that discussed for the cervical spine. Routine AP and lateral plain radiographs constitute the initial evaluation of the thoracic spine, with the exception of the upper thoracic vertebrae (discussed separately later in this chapter). Unilateral or bilateral focal bulging of the thoracic paraspinous soft tissue shadows (the mediastinal stripe or paraspinal line) is an important marker of subtle thoracic fractures (Fig. 2624). Alteration in the contour of the paraspinous shadow is not specic for hematoma unless there is an appropriate history and corresponding ndings on physical examination. Paraspinous abscess or
722
SECTION II Spine
Print Graphic
Presentation
FIGURE 2624. Paraspinal hematomas indicating thoracic spine fracture. A, The anteroposterior coned-down view of the lower thoracic spine shows bulging of the paraspinal stripes (arrowheads) accompanied by a widened disc space at T11T12 (open arrow). B, The coned-down lateral projection shows hyperextension injury with partial inferior end-plate avulsion from T11 (arrow).
neoplasm can produce a soft tissue density similar to that caused by a localized traumatic hematoma. The great majority of acute thoracic spine injuries are recognizable on the initial plain lm examination, and it is not as frequently necessary to use CT to establish the primary diagnosis as it is in the cervical spine. The only segment of the thoracic spine that requires special attention with plain radiography is the cervicothoracic junction. It is essential to be aware that the upper four or ve thoracic vertebrae are not routinely visible on lateral radiographs of the thoracic spine because of the density of the superimposed shoulders. It is therefore incumbent on the attending physician to indicate specically to the radiologist when the upper thoracic spine is the area of suspected injury so that additional views can be obtained. The AP view of both the thoracic and the lumbar spine is also quite helpful in plain lm evaluation. Vertebral alignment is assessed using the position of the pedicles, the presence or absence of scoliosis, and alignment of the spinous processes. The architecture of the pedicles at the affected level is important, as the relationship between spinous processes can be helpful in demonstrating ligament disruption. A sudden increase in distance between two adjacent spinous processes, as determined from the AP lm, is frequently associated with disruption of the intraspinous and supraspinous ligaments and of the facet capsules. Lateral translation on the AP view, combined with anterior translation on the lateral view, suggests a grossly unstable shearing injury. When the patients condition permits, the Fletcher view provides an off-lateral projection of the upper thoracic segments. Positioning for the Fletcher view requires that the patient be rotated slightly from the true lateral, with
one shoulder anterior and the other posterior to the spine. Thus, the upper thoracic segments are projected in slight obliquity between the rotated shoulders. True lateral views of the upper thoracic spine may require CT with sagittal re-formation. Supine oblique views, which can be invaluable in the evaluation of the lower cervical region, are of little value in the upper thoracic spine because of superimposition of the ribs and the complexity of the costovertebral articulations.
Plain Film Radiography of the Lumbar Spine

Radiographic evaluation of the acutely injured lumbar spine, like that of other segments of the spinal column, begins with AP and lateral plain radiographs. When it is clinically inappropriate to place the patient in the true lateral position, the lateral examination should be carried out using a horizontal beam with the patient recumbent. For patients with a history of acute trauma, the lateral spot radiograph of the lumbosacral junction is neither indicated nor necessary. On initial evaluation, the overall alignment of the thoracolumbar junction and lumbar spine is most clearly assessed with a lateral radiograph taken in the supine position. Many fractures demonstrate not only a comminution of the vertebral body but also a local area of kyphosis. The complete loss of lumbar lordosis in the absence of obvious pathology may still be suggestive of injury. As in the cervical spine, subtle rotational injuries are often evident on the lateral projection. In
723
burst fractures of the lumbar spine, the degree of canal compromise can frequently be estimated by observing the posterosuperior corner of the injured vertebral body. The AP view provides the same information as described previously for the thoracic spine. Oblique projections of the lumbar spine should be obtained only when the AP and lateral radiographs are grossly negative and inconsistent with the clinical evaluation. Also, the patients condition must allow rotation into the oblique position. The oblique projection provides another perspective of the lumbar vertebral body and excellent visualization of the pars interarticularis and the facet joints.
Plain Radiography of the Sacrum and Coccyx

Acute injuries involving the sacrum are most commonly associated with pelvic ring disruption. However, isolated injuries of the lower sacral segments and coccyx do occur and require special imaging techniques. The sacral and coccygeal concavity makes adequate visualization of all these segments on a single AP projection impossible. Consequently, in addition to the straight AP radiograph, standard plain radiographic examination of the sacrum and coccyx must include rostrally and caudally angulated AP views as well as a true lateral projection. Superimposition of intestinal artifacts, pelvic calcications, and soft tissue structures can obscure minimally displaced fractures of the sacral and coccygeal segments in the AP projection. In such a case, the fracture is usually evident on the lateral radiograph. CT may be required to detect subtle injuries not evident with radiography. Sacrococcygeal dislocation, even when grossly displaced, is difcult to diagnose radiographically because of the range of normal variation at this level and the effects of pelvic delivery in women. Clinical correlation is particularly important for these patients.
DIAGNOSTIC MODALITIES IN IMAGING SPINAL TRAUMA: ADVANTAGES AND LIMITATIONS
Computed Tomography of the Spine

CT allows images to be obtained in any plane determined by the radiologist to demonstrate the pathology in question to maximal advantage. Multiplanar computed tomography is CT with routinely obtained sagittal and coronal reformatted images. The role of multiplanar CT in the evaluation of injuries of the axial skeleton has been well established.* Simply put, multiplanar CT (including three-dimensional CT) is currently the imaging technique of choice for spinal injury. The principal value of CT is in the axial image, which demonstrates the neural canal and the relationship of
*See references 1, 3, 7, 35, 36, 54, 58, 87, 92, 115, 116.
fracture fragments to the canal. Axial data obtained in the supine patient are converted electronically into images displayed in the sagittal and coronal planes, without requiring movement of the injured patient. The development of multislice CT technology with 0.5-second gantry rotation allows up to eight axial images to be acquired per second and is expected to continue to expand to more images per second in the near future. The speed of data acquisition decreases the patients motion and permits thinner section images to be routinely obtained than with single-slice spiral CT. These factors contribute to a major improvement in the quality of reformatted twodimensional (2-D) and three-dimensional (3-D) images. The volume elements obtained (voxels) with multislice spiral scanning can be made equivalent in size in all three orthogonal axes (isotropic), permitting image quality equivalent to that of axial images in any orientation. Addition of more detector arrays is anticipated to lead to further increases in the speed of image acquisition and improvements in image quality. Spinal CT imaging is performed without intrathecal contrast to (1) evaluate uncertain radiologic ndings, (2) provide details of osseous injury as an aid to surgical planning, (3) assess focal or diffuse spine pain when no radiologic abnormalities are demonstrated, (4) clear the lower cervicothoracic region in symptomatic patients in whom cervical radiography provides inadequate visualization, (5) assess the adequacy of internal xation and detect postoperative complications, and (6) localize foreign bodies and bone fragments in relation to neural elements. CT imaging is not indicated for some spinal injuries identied radiographically. These include simple wedge compression, clay-shovelers fracture, anterior subluxation of the cervical spine, hyperextension teardrop fracture,38 typical hangmans fractures, and typical odontoid fractures. In the thoracic, lumbar, and sacrococcygeal spine, CT is used primarily to assess the relationship of bone fragments to the neural canal, localize penetrating foreign bodies, record the details of complex fracture patterns, and exclude osseous injury with greater accuracy when plain lms are negative in symptomatic patients.45 Ballock and colleagues4 showed that CT is more accurate than plain radiography in distinguishing wedge compression fractures from burst fractures in the thoracolumbar spine. CT often shows additional injuries not suspected on review of plain radiographic views.74 CT is particularly useful in identifying fractures of the occipital condyles22 (Fig. 2625), articular mass and laminae that may occur in association with hyperexion facet dislocations134 (Fig. 2626), hyperextension fracture-dislocations, hyperexion teardrop fractures (Fig. 2627), axial loading fractures of C1 (Jefferson fracture) (Fig. 2628), and vertebral body burst fractures (Fig. 2629). Assessment of subluxation and dislocation is aided by two-dimensional multiplanar as well as surface contour 3-D image re-formation (Figs. 2630 and 2631). The quality of both two-dimensional re-formations and 3-D surface contour images is improved by the use of thinner axial CT slices and by overlapping axial CT images. Axial CT slice thickness should be no greater than 3 mm in the
724
SECTION II Spine
Computed Tomography with Intrathecal Contrast

CTM is multiplanar CT performed after the intrathecal introduction of nonionic water-soluble contrast medium. Depending on the patients condition and the level of suspected cord involvement, the contrast medium can be introduced in the usual myelographic fashion or, more often, laterally at the C1-C2 level with the patient supine.58 Because it is nonviscid, nonionic contrast medium can be introduced through a 22-gauge needle, which can then be removed, and absorbed through the meninges and subarachnoid villi. The water-soluble contrast medium diffuses through the cerebrospinal uid; as a result, less movement of the patient is required to visualize
Print Graphic
Presentation
FIGURE 2625. Computed tomography (CT) scans of occipital condyle fracture. A, Axial CT image through the occipital condyles shows a minimally displaced vertical fracture through the right occipital condyle (arrow) caused by axial loading. B, Coronal plane re-formation shows the fracture crossing vertically through the right condyle (arrow). (A, B, From Mirvis, S.E.; Young, J.W.R. In: Mirvis, S.E.; Young, J.W.R., eds. Imaging in Trauma and Critical Care. Baltimore, Williams & Wilkins, 1992, p. 291.)
Print Graphic
Presentation
cervical spine and 5 mm in the thoracic and lumbar spine and can be routinely made thinner using multislice acquisition scanners. Spiral CT scanners allow reconstruction of images at any slice thickness down to 0.5 mm and therefore generally provide higher-quality reformatted images. Use of such thin-section scans and slice overlap assists in detection of fractures that are oriented in the plane of scanning (axial) such as the type II or low odontoid fracture as well as any minimally displaced fracture. Potential limitations of axial CT include volume averaging (accentuated by use of thick and nonoverlapping axial images that may simulate or obscure a fracture, particularly those oriented along the axial imaging plane); radiation exposure; and time constraints. Minimally displaced fractures may be difcult to identify on reformatted sagittal and coronal images because of degradation in spatial resolution inherent in these images. CT quality is also adversely affected by motion of the patient. As described earlier, multidetector spiral CT with 0.5-second tube rotation will signicantly diminish these current CT limitations.
FIGURE 2626. Computed tomography (CT) scans of unilateral facet fracture-subluxation aided by multiplanar re-formation (MPR). A, Axial CT image shows a complex fracture involving the C5 right articular mass and a posterior avulsion fracture of the C5 vertebral body cortex associated with a rotational injury. B, The two-dimensional (2-D) re-formation in the sagittal plane conrms that the superior articular process of C6 (asterisk) has fractured into and vertically split the C5 right lateral mass, accounting for the complex axial CT image. Vertical splitting fractures of the articular mass are not uncommon in unilateral facet fracture-subluxations. (A, B, From Shanmuganathan, K.; Mirvis, S.E.; Levine, A.L. AJR 163:1165, 1994.)
725
Print Graphic
Presentation
FIGURE 2627. Computed tomography (CT) of hyperexion teardrop fracture. A, Lateral cervical radiograph shows a triangular, anteriorly displaced fragment at C5 with retrolisthesis of the C5 body. Posterior elements appear intact. B, Axial CT image reveals that the fracture has three-column involvement, with two fractures in the laminae. A vertical splitting fracture of C5 is observed in addition to anterior compressions, and there is diastasis of the right facet articulation (arrow). In certain complex fracture patterns, CT is far more useful than plain radiography to elucidate the spectrum of injuries.
the clinically indicated areas of the spine.58 Typically, the area of interest is examined uoroscopically with spot and overhead radiography. If desired, CTM can be performed immediately after introduction of nonionic contrast medium, as current CT scanners can easily produce goodquality images despite the high density of the contrast material.16, 23, 77, 84, 157 CTM provides direct visualization of the spinal cord, cauda equina, and nerve roots, thereby permitting distinction between extramedullary and intramedullary cord or
root injury, localization of cord compression by fracture fragments or herniated disc, or identication of root avulsion,141 partial or complete block of cerebrospinal uid,92 dural tear,104 or post-traumatic syringomyelia.73 The presence of contrast medium in the cord itself indicates a penetrating injury, such as might be caused by a fracture fragment displaced into the canal.31 Many applications of CTM for the evaluation of spinal cord injury have been replaced by MRI. In the institutions in which magnetic eldcompatible immobilization, support, and monitoring systems are available, MRI of the spinal cord should be performed as soon as clinically feasible in all patients with myelopathy. If MRI is not available, the traditional indications for CTM remain valid. CTM, however, remains the imaging technique of choice for demonstrating the presence and extent of dural tears, nerve root herniation, and root avulsion31 (Fig. 2632).
Print Graphic
Computed Tomography with Two-Dimensional Multiplanar and Three-Dimensional Re-formation

3-D CT is the logical extension of the concept of sagittal and coronal re-formation of axial image data. In essence, 3-D CT software programs transform axial CT data into a 3-D optical illusion of the portion of the spinal skeleton being examined. The 3-D images are derived from the data of the axial CT scan.50, 64, 148 Consequently, 3-D reformation increases neither examination time nor radiation dose to the patient. In more contemporary CT systems, the CT data are transferred immediately upon acquisition to independent workstations dedicated to manipulation of large image
Presentation
FIGURE 2628. Computed tomographic scan of a Jefferson burst fracture. Axial image through the C1 level shows a ve-part fracture of the C1 ring. Fractures are minimally displaced, making plain radiographic diagnosis more difcult.
726
SECTION II Spine
Print Graphic
Presentation
FIGURE 2629. Imaging of vertebral burst fracture. A, Lateral cervical radiograph shows loss of height of the C7 body, indicating compression of the anterior and posterior cortices. B, The extent of the injury is better seen by CT, which shows a signicant retropulsed fragment (arrow) and a left lamina fracture (arrowhead).
data sets. Some workstations can be programmed to present instant preselected 3-D renderings of the spine with or without adding surrounding soft tissues. Tissues of different density are assigned different colors to enhance distinction. The 3-D volume images can then be manipulated in real time to nd the preferred angle of viewing or perspective to enhance appreciation of pathology.37 The spine can be electronically cut along any axis to view the neural canal from within or to delete anatomic structures that might obscure the skeletal pathology. The surface contourrendered 3-D image clearly denes and reduces or eliminates ambiguity of complex fractures and fracture-dislocations (Fig. 2633; see also Fig. 2631).
Print Graphic
Magnetic Resonance Imaging of the Spine

Simplistically, MRI scans are derived from the energy released by the hydrogen protons of the body. When placed in a magnetic eld, these protons change their orientation and energy state because of an additional radio-frequency current introduced into the static uniform external magnetic eld. Only tissues within a specic slice within the body have protons precessing at the correct frequency to absorb the radio-frequency energy. Release of this excess applied radio-frequency energy (relaxation) accompanies reorientation of hydrogen protons with the external magnetic eld. A map is created of the location and energy intensity at each point in a slice that reects the magnetic properties of the particular tissues within the slice. MR images are also inuenced by the number of protons within a tissue relative to other tissues, the bulk and microscopic movement of protons, and the chemical state of some tissues such as hemoglobin. Contrast agents such as gadolinium chelates can be used to manipulate tissue relaxation properties and increase or decrease signal and therefore intensity. The intrinsic advantages of MRI have made it the
Presentation
FIGURE 2630. Computed tomographymultiplanar re-formation. A, The axial image shows reversal of the normal alignment of the left articular facets at C4C5 (arrow) and rotation of the vertebral body. B, A two-dimensional re-formation of the axial images in the sagittal plane shows the inferior articular process of C4 locked anterior to the superior articular process of C5 (arrow).
727
Print Graphic
Presentation
FIGURE 2631. Three-dimensional (3-D) computed tomographic surface contour rendering of complex injury. 3-D surface contour views from above (A) and anterior perspective (B) show a complex type II odontoid fracture. The 3-D image improves appreciation of the posterior and lateral translation of the atlas and cranium relative to the C2 body and pronounced compromise of the cervical spinal canal. (A, B, From Mirvis, S.E.; Young, J.W.R. In: Mirvis, S.E.; Young, J.W.R., eds. Imaging in Trauma and Critical Care. Baltimore, Williams & Wilkins, 1992, p. 369.)
imaging technique of choice for the central nervous system, including the spinal cord, its meninges, and its roots.5, 53, 81, 98, 99, 101, 102, 103, 114, 144 These advantages include 1. Direct imaging of the spine in any orientation 2. Superior contrast resolution, when compared with other techniques, in the detection of soft tissue injury, including ligaments, with greater sensitivity 3. Creation of myelography-equivalent images to assess the epidural space for evidence of hematoma, bone fragments, herniated disc material, and osteo-
phytes without use of instilled intrathecal contrast medium 4. Direct imaging of the spinal cord to detect evidence of contusion, hematoma, or laceration 5. Provision of prognostic information regarding the potential for recovery of function based on the MRI appearance of cord injuries 6. Visualization of owing bloodwhich appears dark or bright, depending on the imaging sequence usedfor assessment of major blood vessels, such as the vertebral arteries, without the necessity for intravascular contrast enhancement
Print Graphic
Presentation
FIGURE 2632. Myelography and computed tomography (CT)myelography for evaluation of cervical nerve root avulsion. A, Anteroposterior view from the cervical myelogram reveals post-traumatic pseudomeningoceles arising from the torn nerve roots at the C7 and T1 levels (arrows). B, Axial CT scan through the C7 level after injection of contrast medium shows a small left pseudomeningocele (arrow).
728
SECTION II Spine
Print Graphic
Presentation
FIGURE 2633. Three-dimensional (3-D) computed tomographic surface contour rendering of complex thoracolumbar spine fracture-dislocation. A, 2-D sagittal re-formation shows L1 compression and a retropulsed fragment into the canal. B and C, Surface contour images reveal a marked degree of lateral translation of the thoracic spine relative to L1, improving appreciation of the total injury pattern.
7. No requirement for intravenous contrast material or ionizing radiation Several imaging sequences are routinely performed to emphasize various aspects of normal and pathologic anatomy. In general, most centers employ the following sequences: 1. Sagittal T1-weighted spin echo to dene basic anatomy (Fig. 2634) 2. Sagittal proton and T2-weighted spin echo sequence to emphasize pathologic processes and ligamentous structures (Fig. 2635) 3. Sagittal gradient echo sequence to optimize detection of hemorrhage and distinguish osteophytes from disc material (Fig. 2636) 4. Axial T1-weighted spin echo to assess the epidural space, spinal cord, and neural foramen through areas of interest seen on sagittal sequences (Fig. 2637)
5. Axial gradient echo sequences to visualize gray-white matter delineation and exiting nerve roots and neural foramen (Fig. 2638) 6. Optional MRA sequence to assess cervical arteries, depending on the type of spinal injury (Fig. 2639) A variety of other sequences are now also available that may improve detection of certain types of spinal pathology. The use of a particular sequence (uid attenuation inversion recovery [FLAIR]) can improve detection of subtle spinal cord contusions compared with other standard imaging sequences (Fig. 2640). The limitations of MRI are few but should be mentioned. Because cortical bone contains essentially no hydrogen atoms, it is not well visualized by MRI. Bone is identied by the proton signal of blood and fat in its medullary portion. Consequently, only major osseous injuries are reliably shown by MRI, and MRI cannot be
729
depended on to diagnose subtle bone injury, particularly that involving the posterior spinal elements.76, 79 Comprehensive MRI examinations of the spine require more time than comparable CT studies because of the longer data acquisition. New imaging acquisition sequences can potentially make MRI as fast as or faster than CT scanning. Hemodynamically unstable patients should not be studied by MRI, as acute cardiopulmonary resuscitation is not easily or safely performed in the MRI environment. The need for sophisticated physiologic monitoring and support requires MRI-compatible systems that can function reliably with the fringe magnetic eld around the MRI device and that do not create radio frequency noise in the image acquisition process. The development of such systems has made the application of MRI to patients with acute spinal injury possible.75, 93, 94, 96, 99, 136 However, patients with ferromagnetic intracranial aneurysm clips and pacemakers are excluded from the MRI environment. Some aneurysm clips undergo torque when moved through the external magnetic eld, and pacemakers can malfunction.136, 137 Also, patients with metal in close
Print Graphic
Presentation
Print Graphic
FIGURE 2635. Magnetic resonance imaging of normal sagittal T2weighted sequence. The sagittal T2-weighted spin echo image shows the cerebrospinal uid (CSF) as bright, surrounding an intermediate-signal cord. The vertebral bodies are less dark than on gradient echo sequences and show less contrast with disc material. The posterior longitudinal ligamentanulus brosus complex (closed arrows) is dark between the disc and the CSF The ligamentum avum (arrowheads) appears dark, . outlined anteriorly by bright CSF The open arrow shows an artifact. (From . Mirvis, S.E.; Ness-Aiver, M. In: Harris, J.H., Jr.; Mirvis, S.E., eds. Radiology of Acute Cervical Spine Trauma. Baltimore, Williams & Wilkins, 1995, p. 143.)
Presentation
FIGURE 2634. Magnetic resonance imaging of normal sagittal T1weighted sequence. On T1 weighting, the cerebrospinal uid (CSF) appears dark and the cord intermediate in signal. The anterior anulus and anterior longitudinal ligament appear as a low-signal band (arrowheads) outlined by brighter signals from the disc material (open arrow) and prevertebral fat. The ligamentum avum appears as a dark signal outlined by brighter signal fat on the posterior aspect (closed arrows). The low-signal posterior anulus brosus and posterior longitudinal ligament are poorly seen because of the dark CSF (From Mirvis, S.E.; Ness-Aiver, . M. In: Harris, J.H., Jr.; Mirvis, S.E., eds. Radiology of Acute Cervical Spine Trauma. Baltimore, Williams & Wilkins, 1995, p. 140.)
proximity to vital soft tissue structures such as the spinal cord, nerve roots, or orbit are at increased risk for further tissue injury when positioned in the magnet, particularly when metal foreign body penetration is acute. Patients for whom a history regarding possible exposure to metal fragments cannot be obtained (e.g., welders who are unconscious) must be screened radiographically for the presence of metal foreign bodies. Finally, approximately 3% of patients are sufciently claustrophobic to preclude their being placed inside the magnet bore.75 MRI is indicated in the evaluation of all patients with incomplete or progressive neurologic decit after cervical spinal injury if permitted by the patients overall clinical status. Patients with complete decits should also undergo MRI assessment to demonstrate any cord-compressing lesions (e.g., herniated disc material, epidural hematoma, or bone fragments), the removal of which may allow some neurologic improvement. Other patients for whom spinal MRI is indicated include those with myelopathy or radiculopathy after spinal trauma but with radiographic or CT studies that are negative or fail to account for the decit. Another strong indication is that the level of the
730
SECTION II Spine
Print Graphic
Print Graphic
Presentation
Presentation
FIGURE 2637. Magnetic resonance imaging of normal axial T1-weighted sequence. The axial T1-weighted image shows an intermediate signal cord surrounded by dark cerebrospinal uid. The nerve roots are seen traversing the subarachnoid space (open white arrows) and exiting the neural foramen (arrowheads). The cortical bone is dark, with brighter signal marrow. The vertebral arteries (open black arrows) are dark in this sequence (ow void). No internal anatomy in the cord is discerned. (From Mirvis, S.E.; Ness-Aiver, M. In: Harris, J.H., Jr.; Mirvis, S.E., eds. Radiology of Acute Cervical Spine Trauma. Baltimore, Williams & Wilkins, 1995, p. 144.)
FIGURE 2636. MRI of normal sagittal gradient echo sequence. The sagittal gradient echo sequence normally shows the vertebral bodies as relatively low in signal (because fat signal decreases). The intervertebral discs and cerebrospinal uid (CSF) remain bright, allowing demonstration of the combined low signal of the anulus brosus and longitudinal ligaments. The intermediate signal intensity of the cord is easily seen, surrounded by bright CSF .
decit does not correlate with the injury location depicted by radiography or CT.101, 102, 107, 124 Finally, MRI can demonstrate the level and extent of ligament disruption and intervertebral disc herniation. This information helps determine the need for and the type of internal xation required to restore a patent spinal canal and ensure mechanical stability.28 For thoracic injuries, MRI is useful in dening the extent of posterior
Print Graphic
Presentation
FIGURE 2638. Magnetic resonance imaging study of normal axial gradient echo sequence. A and B, Gradient echo images produce very dark bone and bright cerebrospinal uid (CSF). The dura is outlined by CSF (arrows in A). Internal architecture of the cord, with brighter central gray and darker white matter tracts, can be observed (A). Vertebral arteries appear bright on this sequence (open arrows in B). Nerve roots can be seen within the neural foramen surrounded by high-signal CSF (white arrows in B). Facet articular spaces contain high-signal uid (arrowheads in B). (A, B, From Mirvis, S.E.; Ness-Aiver, M. In: Harris, J.H., Jr.; Mirvis, S.E., eds. Radiology of Acute Cervical Spine Trauma. Baltimore, Williams & Wilkins, 1995, p. 145.)
731
Print Graphic
Presentation
FIGURE 2639. Magnetic resonance angiography (MRA) study of normal time-of-ight cervical vessels. MRA study demonstrates anteroposterior (AP) (A) and oblique (B) views of the cervical vasculature. All vessels appear bright on the gradient echo sequence used to acquire images. The venous ow signal is selectively negated. Note the bright signal from moving cerebrospinal uid in the AP projection. (A, B, From Mirvis, S.E.; Ness-Aiver, M. In: Harris, J.H., Jr.; Mirvis, S.E., eds. Radiology of Acute Cervical Spine Trauma. Baltimore, Williams & Wilkins, 1995, p. 146.)
FIGURE 2640. Increased sensitivity for cord edema with inversion recovery sequence. A, Lateral T2-weighted spin echo sequence shows mild spinal stenosis of C4-5 but no cord lesion. B, Inversion recovery sequence shows central cord signal increase from C2-3 to mid C6 compatible with neurologic decit (arrows).
Print Graphic
Presentation
732
SECTION II Spine
ligament injury, which increases instability when associated with anterior column fractures. MRI is particularly helpful in determining the extent of ligament injury and instability that typically accompany injuries occurring in the fused spine, such as ankylosing spondylitis and diffuse idiopathic skeletal hyperostosis. MAGNETIC RESONANCE IMAGING OF SPECIFIC ACUTE SPINE INJURY Parenchymal Lesions MRI is unique in its ability to detect acute injury to the spinal cord, including edema, hemorrhage, and laceration. Cord edema appears isointense or slightly hypointense in relation to the normal cord on T1-weighted spin echo images but becomes brighter in signal than the normal cord on T2-weighted image sequences (Fig. 2641). When hemorrhage is present within the cord, its MRI appearance depends on a complex relationship between the chemical state of the blood, the eld strength of the magnet, and the imaging sequence used.14 In the acute to subacute period after injury (1 to 7 days), blood generally appears dark (low-intensity signal) on T2-weighted sequences, whereas edema has a bright signal (Fig. 2642). After about 7 days, as red cells are lysed, blood acquires a high-intensity signal in both T1- and T2-weighted studies.
Kulkarni and colleagues81 were the rst to describe a relation between the characteristics of the MRI cord signal and the patients outcome, suggesting that MRI cord signal characteristics reect the type of cord histopathology that is, hemorrhage (type 1), edema (type 2), and mixed edema and hemorrhage (type 3). The prognostic information provided by MRI regarding potential recovery of function has been veried by several other studies.19, 41, 90, 98, 129, 144 The ability of the MRI signals to identify the histopathology of acute cord injury has been conrmed by direct comparison of the MRI signal with histologic ndings in experimentally induced spinal cord injuries.19, 120, 153 Ligament Injury Ligament injury sustained in acute spinal trauma is inferred from the mechanism of injury, the ultimate fracture pattern, and the alignment of the spine after injury. However, even signicant ligament injury leading to spinal mechanical instability, particularly hyperexion and hyperextension sprains without concurrent fractures, may not be apparent when the spine is studied radiographically in the neutral position. Furthermore, the spinal alignment demonstrated by plain radiographs may serve to reveal the site of major or principal mechanical instability but may not demonstrate all major ligament injuries and other potential sites of immediate or delayed instability. MRI depicts normal ligaments as regions of low signal intensity because of lack of mobile hydrogen. Disruption of the ligament is seen on MRI scans as an abrupt interruption of the low signal, ligament attenuation or stretching of the ligament, or association of a torn ligament with an attached avulsed bone fragment (Figs. 2643 to 2645; see also Fig. 2641). Determination of the status of the major support ligaments of the spine as revealed by MRI has a denite bearing on management approaches.17, 18, 32, 78, 138 MRI can demonstrate unsuspected ligament injury or injury that is greater than anticipated from the results of other available imaging modalities.151 Intervertebral Disc Herniation Acute intervertebral disc herniation may accompany fractures or dislocations or may occur as an isolated lesion. If the disc impinges on the spinal cord or roots, a neurologic injury may result. MRI demonstration of a single-level acute intervertebral disc herniation that impinges on the spinal cord is crucial in surgical management of spinal trauma to optimize neurologic recovery. MRI clearly depicts disc material herniation with essentially all imaging sequences (Fig. 2646) but best separates disc material from posterior osteophyte with the gradient echo sequence, on which relatively bright disc material is visualized against a dark background of bone. The advantage of MRI over CTM in detecting acute traumatic disc herniation was shown clearly by Flanders and associates.41 In their study, 40% of acute disc herniations producing neurologic decits were demonstrated by MRI but not by CTM. Rizzolo and colleagues124 found a 42% incidence of herniated nucleus pulposus in 53 patients studied by MRI at 1.5 T within 72 hours of injury. The highest incidence occurred among patients
Print Graphic
Presentation
FIGURE 2641. Magnetic resonance imaging scan of cord edema. Sagittal T2-weighted spin echo study shows area of increased signal in the spinal cord at the C4C5 level representing focal cord edema (arrowhead). There is disruption of the adjacent low-signal ligamentum avum, indicating disruption from hyperexion (arrow).
733
Print Graphic
Presentation
FIGURE 2642. Magnetic resonance imaging of cord hemorrhage. A, Lateral cervical spine radiograph of blunt trauma victim reveals hyperexion teardrop fracture of C6 (arrow). There is also a compression fracture of the anterior superior endplate of C5. There is prevertebral soft tissue swelling. B, A lateral sagittal T2 spin echo image reveals edema (high signal) extending from middle C3 to C7. Focal areas of low signal represent hemorrhage. Edema is seen as a bright signal in C5 and C6 vertebrae. C, Gradient echo axial image reveals a focal low-signal area in the left side of the spinal cord (arrow). The low signal results from hemorrhage consisting of intracellular methemoglobin and deoxyhemoglobin. The higher signal area in the adjacent cord is due to edema. A fracture is seen in the vertebral body.
with bilateral facet dislocations (80%) and anterior cord syndromes (100%). Doran and co-workers34 described a high incidence of traumatic disc herniation among patients with both unilateral and bilateral facet dislocations. Patients with traumatically herniated intervertebral discs may sustain a neurologic deterioration when the cervical spine is reduced, as the disc may then compress neural tissue.11, 34, 53, 118 However, this point is controversial, as others nd no evidence of neurologic deterioration when closed reduction is performed for patients with disc herniation or disruption.50
Epidural Hematoma Epidural hematomas (EDHs) are an uncommon sequela of spinal trauma and occur in 1% to 2% of cervical spine injuries.44 The cervical spine is the most common location of EDHs of traumatic origin.44 EDH most commonly occurs in the dorsal epidural space as a result of close adherence of the ventral dura to the posterior longitudinal ligament.44 Bleeding most likely arises from sudden increases in pressure in the rich epidural venous plexus, which comprises valveless veins.44, 109 EDHs may develop
734
SECTION II Spine
acutely after trauma, in a delayed fashion, or after open or closed spinal column reduction. Up to 50% of posttraumatic EDHs may occur among patients without overt cervical spine injuries.44 For this reason, the presence of myelopathy without an injury demonstrated by radiography or CT without intrathecal contrast should suggest an EDH. Garza-Mercado4 described an increased likelihood of cervical spine EDH in younger trauma victims owing to increased elasticity of the vertebral column and among patients with fused cervical spines, including those with ankylosing spondylitis and diffuse idiopathic skeletal hyperostosis. The development of progressive, unexplained neurologic deterioration among patients sustaining spinal trauma may herald the onset of cord compression by an expanding EDH. Again, the MRI appearance of EDH depends on the age of the blood, magnetic eld strength, and imaging sequence used. In the acute phase of trauma (1 to 3 days after injury), blood appears isointense (bright) relative to the spinal cord on the T1-weighted sequence (Fig. 2647) and hypointense (dark) relative to the spinal cord on T2-weighted sequences. At 3 to 7 days after injury, the central portion of the hematoma, which contains intact red blood cells, has low signal intensity on T2-weighted sequences, whereas the periphery, composed of lysed red blood cells, shows increased signal strength on both T1and T2-weighted sequences.14 Congenital or Acquired Spinal Stenosis Spinal cord injury may be caused by impaction of posteriorly projecting osteophytes or hypertrophied, calcied, or ossied ligaments on the anterior surface of the cord during traumatic deformation of the cervical spine. Posterior spinal cord injury can result from buckling of hypertrophied ligamentum avum during hyperextension.122 Patients with congenital spinal stenosis or spinal stenosis acquired from degenerative changes (spondylosis) have an increased likelihood of injury from cervical spine
Print Graphic
Presentation
FIGURE 2644. Magnetic resonance imaging study of ligament injury. Sagittal T2-weighted spin echo scan in a quadriplegic trauma patient reveals interruption of the posterior longitudinal ligament and posterior anulus (arrowhead) and the ligamentum avum (arrow). Prevertebral edema (open arrows) is observed. The anterior longitudinal ligament appears intact, indicating hyperexion subluxation mechanism. (From Mirvis, S.E.; Ness-Aiver, M. In: Harris, J.H., Jr.; Mirvis, S.E., eds. Radiology of Acute Cervical Spine Trauma. Baltimore, Williams & Wilkins, 1995, p. 151.)
Print Graphic
Presentation
FIGURE 2643. Magnetic resonance imaging of ligament tear. Axial T2-weighted image through the C2 level in a trauma patient shows interruption of the transverse atlantal ligament (arrow) and displacement of the odontoid process to the left. (From Mirvis, S.E.; Ness-Aiver, M. In: Harris, J.H., Jr.; Mirvis, S.E., eds. Radiology of Acute Cervical Spine Trauma. Baltimore, Williams & Wilkins, 1995, p. 156.)
trauma or even physiologic cervical spine motion. The occurrence of post-traumatic myelopathy without radiologic evidence of acute injury among older patients with posterior spinal osteophytes, ossication of the posterior longitudinal ligament, or congenital spine stenosis suggests that these conditions are etiologic.80, 145 Cervical spinal cord impaction by posterior cervical osteophytes typically produces a central cord syndrome.20 MRI in the sagittal and axial orientation depicts spinal canal compromise produced by degenerative processes (Fig. 2648). Comparison of T2-weighted spin echo and T2-weighted gradient echo sequences can be helpful in differentiating acutely herniated soft disc material from osteophytes surrounding chronic disc herniations. Both sequences produce a myelographic appearance that demonstrates the relationship of osteophytes and intervertebral discs to the spinal cord. However, gradient echo sequences produce very dark-appearing osteophytes and increased contrast with brighter signal disc material compared with these features on T2-weighted spin echo sequences. MRI is crucial in planning the extent of posterior surgical decompression by showing the points at which the thecal sac and direct spinal cord compression occur. It should be
735
Print Graphic
FIGURE 2645. Magnetic resonance imaging of ligament disruption. A, Lateral lower thoracic spine radiograph shows widening anteriorly in the disc space between two thoracic vertebrae (arrow). B and C, Sagittal T2-weighted spin echo sequence shows minimal bright hemorrhage just above the disc material and complete interruption of the dark (low-signal) anterior anulus brosus bers and longitudinal ligament (arrow).
Presentation
736
SECTION II Spine
Print Graphic
Print Graphic
Presentation
Presentation
FIGURE 2646. Magnetic resonance imaging study of disc herniation. Off-midline sagittal proton density scan shows chronic disc herniation at the C6C7 level indenting the spinal cord. The presence of low-signal osteophytes around the disc suggests chronicity, possibly with acute exacerbation after trauma.
FIGURE 2648. Magnetic resonance imaging scan of cord contusion secondary to spinal stenosis. The sagittal T2-weighted spin echo image in a trauma patient with central cord syndrome shows marked narrowing of the spinal canal at the bottom of C3 to C5C6 because of bulging discs and hypertrophic ligamentum avum. A cord contusion at C3 and C4 is evident as increased signal from edema (arrowheads).
Print Graphic
Presentation
FIGURE 2647. Magnetic resonance imaging (MRI) study of epidural hematoma (acute). Lateral (A) and axial (B) T1-weighted MRI scans of a patient with cervical exion injury shows an epidural hematoma (arrowheads) isointense with the spinal cord, displacing the cord posteriorly. There is C5 on C6 anterior subluxation. (A, B, From Mirvis, S.E.; Ness-Aiver, M. In: Harris, J.H., Jr.; Mirvis, S.E., eds. Radiology of Acute Cervical Spine Trauma. Baltimore, Williams & Wilkins, 1995, p. 160.)
737
noted that the gradient echo pulse sequence tends to make bone appear larger than in actuality (blooming), and this may accentuate the apparent degree of spinal canal encroachment. MAGNETIC RESONANCE IMAGING OF CHRONIC AND POSTOPERATIVE SPINAL INJURIES It has been well documented that MRI is superior to myelography, CT, and CTM in evaluation of chronic injuries of the spinal cord, particularly for the differentiation of myelomalacia and post-traumatic spinal cord cyst.113, 121, 143 Myelomalacia typically appears as a focal low-signal area on T1-weighted sequences and as a high-signal area on T2-weighted sequences in a cord of normal or decreased caliber18 (Fig. 2649). Syringomyelia has a similar appearance but is more sharply delineated and typically occurs in an expanded cord. Flow-sensitive imaging sequences may help to demonstrate a syrinx by demonstrating cerebrospinal uid movement within the cavity.18 Postoperative MRI studies of patients with internal xation devices are improved when titanium xation devices are used.100 These produce far less magnetic susceptibility artifact than stainless steel xation
Print Graphic
Presentation
FIGURE 2650. Magnetic resonance imaging (MRI) study of stainless steel wire internal xation. Sagittal gradient echo MRI scan after stainless steel wire posterior xation shows marked artifact surrounding the wire, obscuring the spinal canal and cord. The increased artifact is due to both the increased magnetic susceptibility of steel wire and the gradient echo sequence, which is signicantly affected by magnetic susceptibility artifact.
Print Graphic
devices and permit visualization of the cord and surrounding epidural space without artifact36, 100 (Figs. 2650 and 2651).
Magnetic Resonance Angiography

Presentation
FIGURE 2649. Magnetic resonance imaging study of chronic spinal cord injury. Sagittal T1-weighted sequence in a patient obtained several months after hyperexion injury (note loss of C7 anterior height) shows irregular low signal in the cord and focal cord atrophy at C6C7, representing myelomalacia. Posterior titanium wire xation produces minimal artifact. (From Mirvis, S.E.; Young, J.W.R. In: Mirvis, S.E.; Young, J.W.R., eds. Imaging in Trauma and Critical Care. Baltimore, Williams & Wilkins, 1992, p. 367.)
MRA is used as a screening assessment of the vertebral arteries. The exact incidence of vertebral artery injury occurring after cervical spine fracture-dislocation is unknown, but the injury is being reported with increasing frequency.* Vertebral artery injuries from cervical spine trauma generally involve the second portion of the artery extending from C6 to C2. Fixation of the artery within the connes of the transverse foramina predisposes this vessel to injury from cervical dislocations. Although a variety of cervical spine injuries have been associated with vertebral artery injury, unilateral and bilateral dislocations are most commonly implicated.112, 130 Vertebral artery injury can occur from fractures extending across the foramen transversarium and has been reported with lateral cervical dislocations.112, 149, 154
*See references 29, 30, 43, 46, 70, 86, 112, 123, 130, 149, 154, 156.
738
SECTION II Spine
MRA screening of the vertebral arteries should be considered for all patients with blunt cervical spine trauma with signicant degrees (>1 cm) of dislocation or subluxation or fracture of the foramen transversarium or with neurologic decits consistent with vertebral vessel insufciency.46, 149, 154, 156 Routine assessment of the cervical spine by MRI should include axial T1-weighted images. On these sequences, owing blood creates a signal void (dark image). Conversely, on gradient echo sequences, owing blood creates a bright image. Inspection of the major cervical arteries for the anticipated signal characteristics should be performed as part of overall assessment of the MRI study. Absence or irregularity of the expected ow signal should raise a question of vascular injury (Fig. 2652). Injuries identiable by MRA include intimal aps, intramural dissection or hematoma, pseudoaneurysm, and thrombosis. Care must be taken to distinguish injury from vessel hypoplasia or atherosclerotic disease. Positive MRA ndings of vessel injury are conrmed and better characterized by direct contrast angiography, which offers spatial resolution greater than that possible with MRA. Penetrating injury accounts for the majority of cervical vertebral injuries resulting from trauma. The presence of retained metal fragments from ballistic injury precludes vascular MRA assessment because of artifacts created by close proximity of metal. In addition, because MRA is less sensitive for detection of subtle intimal injuries or mural hematoma, conventional arteriography is in general recommended for evaluation of suspected vertebral artery injury caused by penetrating force. Although the incidence of vertebral artery injury from blunt trauma to the cervical
spine appears to be higher than previously suspected, the injury usually results in complete thrombosis without producing a neurologic deterioration.30, 46, 154, 156 It also appears that thrombosed vessels remain occluded on long-term follow-up without the need to perform endovascular occlusion. Vertebral arteries that are injured but patent can lead to formation of clot and embolization with infarction. These injuries, when identied, require treatment to prevent or minimize the chance of embolization using antiplatelet or anticoagulant treatment as permitted by the patients condition or open surgical or endovascular treatment.
Catheter Angiography
Conventional angiography is used to detect or conrm vertebral artery injury resulting from cervical spine trauma. Although associated with higher procedurerelated morbidity than MRA, the technique offers greater spatial resolution for detection and characterization of vascular injuries (Fig. 2653). As stated earlier, conventional angiography is the current study of choice for assessment of potential vertebral artery injury resulting from penetrating injury to the cervical spine. In addition, angiography can provide the potential for intravascular thrombolysis and endovascular stent treatment of selected vertebral injuries.119
Radionuclide Bone Imaging

Radionuclide bone imaging (RNBI) has been used in the assessment of trauma to the spine primarily to determine whether a radiographic abnormality represents an acute process that is potentially responsible for the patients pain or to exclude an osseous abnormality as a source of spine pain when radiographs are normal. In the cervical spine, image resolution is improved by placing the patients neck directly on the collimator surface, decreasing distance from the nuclide activity. Slightly posterior oblique images of the cervical spine can also assist diagnostically. Reports91, 140 and anecdotal experience indicate that an acute, nondisplaced cervical fracture cannot be entirely excluded even when the initial RNBI scan is normal. RNBI was assessed in patients with whiplash,6, 66 and no correlation was found between symptoms and signs of injury and scintigraphic ndings. However, one retrospective study of 35 cases6 found that a negative bone scan excluded a skeletal injury, and in another prospective study of 20 patients66 with whiplash injuries, no patients had bone scan ndings suggestive of fracture and none had a subsequent diagnosis of fracture. Increased activity within the cervical spine on delayed bone imaging includes a differential diagnosis of nonspecic stress response, degenerative arthritis, or healing fracture. Use of single photon emission computed tomography may increase diagnostic accuracy in bone imaging of acute spine trauma. RNBI in the thoracic and lumbar spine is technically
Print Graphic
Presentation
FIGURE 2651. Magnetic resonance imaging (MRI) study of titanium implant internal xation. Sagittal T2-weighted MRI scan shows a titanium xation plate at C4C6 used to x a hyperexion teardrop fracture at C5, producing local magnetic eld inhomogeneity and signal dropout. However, the spinal canal and cord are still well seen, with a focal area of post-traumatic cyst formation or myelomalacia visible at the C5 level. There is minimal retrolisthesis of C5.
739
Print Graphic
Presentation
FIGURE 2652. Imaging of vertebral artery injury. A, Axial T1-weighted magnetic resonance imaging scan reveals a lack of expected ow void in the right vertebral artery foramen (arrow), whereas left vertebral ow void is observed (arrowhead). B, Magnetic resonance angiography scan in anteroposterior view shows absence of the right vertebral artery signal, but the left signal is present (arrows). C, Right subclavian angiogram reveals thrombosis (arrow) of the vertebral artery that resulted from unilateral facet dislocation. (AC, From Mirvis, S.E.; Ness-Aiver, M. In: Harris, J.H., Jr.; Mirvis, S.E., eds. Radiology of Acute Cervical Spine Trauma. Baltimore, Williams & Wilkins, 1995, p. 175.)
easier than in the cervical spine. Acute fractures can be detected on both blood pool and delayed images (Fig. 2654). Increased linear activity at the superior end-plate is characteristic of traumatic fracture. RNBI may be particularly helpful in detecting acute compression fractures in patients with severe osteoporosis that may be quite subtle radiographically. Increased lateral activity on the
concave side of a scoliotic spine that is not sharply marginated most likely represents stress-related or degenerative change. In patients with nonlocalized lower back pain after trauma and normal lumbar radiographs, largeeld-of-view RNBI can screen for small laminar, transverse process or articular process fractures that might otherwise require multilevel CT scanning to detect.
740
SECTION II Spine
Print Graphic
Presentation
FIGURE 2653. Angiography of vertebral injury. A, Anteroposterior view of the cervical spine obtained from a digital subtraction angiogram shows a deformed bullet overlying the cervical spine at the left C5C6 articular masses. B, Image from a selective left vertebral angiogram shows a thrombosed (arrow) left vertebral artery below the level of the bullet (the bullets density has been subtracted from the image).
Print Graphic
Presentation
FIGURE 2654. Nuclear scintigraphy of a lumbar compression fracture. A, Lateral lumbar spine radiograph in a trauma patient with mild back pain shows possible L3 compression fracture (X) versus Schmorls node deforming the superior end-plate. B, Static images from bone scintigraphy show end-plate increased tracer activity, indicating acute fracture.
741
IMAGING SPINAL TRAUMA IN PREEXISTING PATHOLOGIC CONDITIONS
zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz Trauma patients with various conditions that lead to fusion of the spine are at increased risk for spinal injury compared with patients with normal spine mobility. Patients with ankylosing spondylitis can sustain spinal injury from minimal amounts of blunt force impact.108 Because the spine has undergone bony ankylosis, it is very fragile and fractures equally easily across the bone or disc spaces (Figs. 2655 and 2656). Usually, these injuries completely traverse all supporting ossied spinal ligaments, creating marked instability. In my experience, these injuries are typically evident in extension or extensiondislocation patterns. It has been noted that spinal fractures in patients with underlying ankylosing spondylitis are not uncommonly occult radiographically.39 Fractures and dislocations most typically occur in the lower cervical followed by the thoracic spine. This nding may be due to spontaneous reduction of the fracture, generalized osteoporosis that often occurs in patients with advanced
Print Graphic
Presentation
FIGURE 2655. Hyperextension injury with ankylosing spondylitis. A, Anterior thoracolumbar radiograph shows apparent increase in height of L3 compared with superior lumbar levels (2-headed arrow). B, Sagittal T1-weighted magnetic resonance image shows fracture across mid-L3 body with posterior displacement and angulation of the superior fragment. The anterior and posterior longitudinal ligaments appear striped away from the adjacent vertebral bodies. Low-signal edema replaces high-signal fat in bone marrow.
disease, and failure to appreciate second, noncontiguous injuries.39 Although patients with ankylosing spondylitis may present with well-established neurologic decits and obvious imaging abnormalities, about one third have a delayed onset of neurologic decits as a result of failure to diagnose or properly immobilize the very unstable spine. In general, patients presenting with blunt trauma who have ankylosing spondylitis should be regarded as having unstable injuries until denitely proved otherwise. If radiographs appear normal, further evaluation by thinsection CT is recommended to detect subtle fractures. CT may also assist in differentiating acute fractures from pseudarthrosis related to previous injury.47 If the patient has cervical pain, further assessment by MRI to detect subtle soft tissue edema or bone marrow edema is also indicated to avoid misdiagnosing a highly unstable injury. Other preexisting conditions that may be associated with an increased risk of spinal fracture with lower energy blunt trauma include diffuse idiopathic skeletal hyperostosis (DISH) or Forestiers disease, also called ankylosing hyperostosis, spinal spondylosis, and osteoporosis. DISH is similar to ankylosing spondylitis in the sense that the spine contains a segment of bone fusion. DISH may be differentiated from ankylosing spondylitis by the absence of squared vertebral body corners; the larger, coarser, and predominantly anterior syndesmophytes of DISH; and lack of sacroiliac and apophyseal changes that occur in ankylosing spondylitis. Fractures in DISH may occur through the midportion of a fused segment or at the top or bottom through a disc space or odontoid process.111 The long lever arm created by the fused segment focuses all the energy of the applied force onto a single disc space, increasing the risk of injury. Similarly, spines that are fused because of multiple contiguous levels of degenerative spondylosis are also at increased risk for injury because of inability to distribute straining forces across multiple spinal levels. Patients with severe osteoporosis are at increased risk for fracture resulting from minor injury or activities of daily living. In the spine, such injuries may appear as minor loss of height of a vertebral body. The age of the injury may not be apparent, and it may be assumed in some cases to be a remote lesion. Compression fractures in patients with structural bone weakness can progress to signicant compression with physiologic loading and produce acute or delayed onset of radicular or complete neurologic decits. The lack of density of demineralized bone associated with suboptimal lm technique renders radiographic interpretation of the spine difcult and insensitive to detection of subtle fractures. If the patient has persistent spinal pain, examination by thin-section CT is recommended initially, as it often demonstrates subtle end-plate fractures and paraspinal hematoma not detected by radiography. If clinical symptoms remain unexplained or CT is not denitive, MRI is suggested. MRI shows paraspinal edema, hematoma, and bone edema with high sensitivity, improving the level of condence in injury detection or exclusion. Nuclear bone scintigraphy can also play a role in diagnosing fractures in this setting. However, acute fractures may not have abnormal bone turnover activity in the acute phase,
742
SECTION II Spine
Print Graphic
Presentation
FIGURE 2656. Hyperextension injury with ankylosing spondylitis. A, Lateral thoracic spine radiograph in blunt trauma patient with ankylosing spondylitis shows widened disc space anteriorly at the T11T12 (arrow) level representing hyperextension injury. B, Proton density sagittal image through the injury site indicates that the fracture crosses obliquely through the T11 vertebral body rather than through the disc space. There is focal spinal cord compression in part caused by posterior bony encroachment and cord edema.
particularly in elderly people. Also, foci of increased nuclide deposition in the spine may be due to chronic abnormalities such as seen with spondylosis or subacute injuries as well as acute pathology, making this examination less useful. In general, very careful attention to imaging for these subsets of patients combined with a low threshold to perform additional diagnostic studies in symptomatic patients is warranted.
REFERENCES 1. Acheson, M.B.; Livingston, R.R.; Richardson, M.L.; et al. Highresolution CT scanning in the evaluation of cervical spine fractures. AJR 148:1179, 1987. 2. Anderson, L.D.; DAlonzo, T.R. Fractures of the odontoid process of the axis. J Bone Joint Surg Am 56:1663, 1974. 3. Angtuaco, E.J.C; Binet, E.F Radiology of thoracic and lumbar . fractures. Clin Orthop 189:43, 1984. 4. Ballock, R.T.; MacKersie, R.; Abitbol, J.; et al. Can burst fractures be predicted from plain radiographs? J Bone Joint Surg Br 74:147, 1992. 5. Banna, M. Clinical Radiology of the Spine and the Spinal Canal. Rockville, MD, Aspen Systems Corporation, 1985, p. 411. 6. Barton, D.; Allen, M.; Findlay, D.; et al. Evaluation of whiplash injuries by technetium 99m isotope scanning. Arch Emerg Med 10:197, 1993. 7. Bauer, R.D.; Errico, T.J.; Waugh, T.R.; Cohen, W. Evaluation and diagnosis of cervical spine injuries: A review of the literature. Cent Nerv Syst Trauma 4:71, 1987. 8. Benzel, E.C.; Hart, B.L.; Ball, P.A.; et al. Fractures of the C2 vertebral body. J Neurosurg 81:206, 1994.
9. Benzel, E.C.; Hart, B.L.; Ball, P.A.; et al. Magnetic resonance imaging for the evaluation of patients with occult cervical spine injury. J Neurosurg 85:824, 1996. 10. Berne, J.D.; Velmahos, G.C.; El-Tawil, Q.; et al. Value of complete cervical helical computed tomographic scanning in the unevaluable blunt patient with multiple injuries: A prospective study. J Trauma 47(5):896, 1999. 11. Berrington, N.R.; Van Staden, J.F Willers, J.G.; et al. Cervical .; intervertebral disc prolapse associated with traumatic facet dislocation. Surg Neurol 40:395, 1993. 12. Bettinger, B.I.; Eisenberg, R.L. Improved swimmers lateral projection of the cervicothoracic region. AJR 164:1303, 1995. 13. Blackmore, C.C.; Ramsey, S.D.; Mann, F .A.; Deyo, R.A. Cervical spine screening with CT in trauma patients: A cost-effectiveness analysis. Radiology 212;117, 1999. 14. Bradley, W.G. MR appearance of hemorrhage in the brain. Radiology 189:15, 1993. 15. Brady, W.J.; Moghtader, J.; Cutcher, D.; et al. ED use of exion-extension cervical spine radiography in the evaluation of blunt trauma. Am J Emerg Med 17:504, 1999. 16. Brant-Zawadzki, M.; Miller, E.M.; Federle, M.P. CT in the evaluation of spine trauma. AJR 136(2):369, 1981. 17. Brightman, R.P.; Miller, C.A.; Rea, G.L.; et al. Magnetic resonance imaging of trauma to the thoracic and lumbar spine: The importance of the posterior longitudinal ligament. Spine 17:541, 1992. 18. Castillo, M.; Harris, J.H., Jr. MRI of the spine: Recent applications. Mediguide Orthop 10(6):1, 1991. 19. Chakeres, D.W.; Flicking, F Bresnahan, J.C.; et al. MR imaging of .; acute spinal cord trauma. AJNR 8:5, 1987. 20. Chang, C.Y.; Wolf, A.L.; Mirvis, S.E.; et al. Body surng accident resulting in cervical spine injuries. Spine 17:257, 1992. 21. Christensen, P.C. The radiologic study of the normal spine. Radiol Clin North Am 15:133, 1977.
CHAPTER 26 Spinal Imaging 22. Clayman, D.A.; Sykes, C.H.; Vines, F Occipital condyle fractures: .S. Clinical presentation and radiologic detection. AJNR 15:1309, 1994. 23. Cooper, P.R.; Cohen, W. Evaluation of cervical spinal cord injuries with metrizamide myelographyCT scanning. J Neurosurg 61:281, 1984. 24. Daffner, R.H. Cervical radiography for trauma patients: A timeeffective technique? AJR 175:1309, 2000. 25. 25 Dai, L; Jia, L. Central cord injury complicating acute cervical disc herniation in trauma. Spine 25:331, 2000. 26. DAlise, M.D.; Benzel, E.C.; Hart, B.L. Magnetic resonance imaging evaluation of the cervical spine in the comatose or obtunded patient. J Neurosurg 91:54, 1999. 27. Davis, J.W.; Parks, S.N.; Detlefs, C.L.; et al. Clearing the cervical spine in obtunded patients: The use of dynamic uoroscopy. J Trauma 39:435, 1995. 28. Davis, S.J.; Teresi, L.M.; Bradley, W.G.; et al. Cervical hyperextension injuries. MR ndings. Radiology 180:245, 1991. 29. Deen, H.G.; McGirr, S.J. Vertebral artery injury associated with cervical spine fractures. Spine 17:230, 1992. 30. Demetriades, D.; Theodorou, D.; Asension, J.; et al. Management options in vertebral artery injuries. Br J Surg 83:83, 1996. 31. Denis, F Burkus, J.K. Diagnosis and treatment of cauda equina .; entrapment in the vertical lumbar burst fracture. Spine 16:S433, 1991. 32. Dickman, C.A.; Mamourian, A.; Sonntag, V.K.; et al. Magnetic resonance imaging of the transverse atlantal ligament for the evaluation of atlantoaxial instability. J Neurosurg 75:221, 1991. 33. Diliberti, T.; Lindsey, R.W. Evaluation of the cervical spine in the emergency setting. Who does not need an x-ray? Orthopedics 15:170, 1992. 34. Doran, S.E.; Papadopoulos, M.; Ducker, T.; et al. Magnetic resonance imaging documentation of coexistent traumatic locked facets of the cervical spine and disc herniation. J Neurosurg 79:341, 1993. 35. Dorwar, R.H.; Lamasters, D.L. Applications of computed tomographic scanning of the cervical spine. Orthop Clin North Am 16:381, 1985. 36. Ebrahaim, N.A.; Rupp, R.E.; Savolaine, E.R.; et al. Use of titanium implants in pedicular screw xation. J Spinal Disord 7:478, 1994. 37. Edeiken-Monroe, B.S.; Wagner, L.K.; Harris, J.H., Jr. Hyperextension dislocation of the cervical spine. AJNR 7:135, 1986. 38. Erb, R.; Schucany, W.G.; Shanmuganathan, K.; et al. Extension corner avulsion fractures of the cervical spine. Emerg Radiol 3:96, 1996. 39. Finkelstein, J.A.; Chapman, J.R.; Mirza, S. Occult vertebral fractures in ankylosing spondylitis. Spinal Cord 37:444, 1999. 40. Fischer, R.P. Cervical radiographic evaluation of alert patients following blunt trauma. Ann Emerg Med 13:905, 1984. 41. Flanders, A.E.; Schaeffer, D.M.; Doan, H.T.; et al. Acute cervical spine trauma: Correlation of MR imaging ndings with degree of neurologic decit. Radiology 177:25, 1990. 42. Flee, C.; Woodring, J.H. Unstable Jefferson variant atlas fracture: An unrecognized cervical injury. AJNR 12:1105, 1992. 43. Gambee, M.J. Vertebral artery thrombosis after spinal injury: Case report. Paraplegia 24:350, 1986. 44. Garza-Mercado, R. Traumatic extradural hematoma of the cervical spine. Neurosurgery 24:410, 1989. 45. Gellad, F .E.; Levine, A.M.; Joslyn, J.N.; et al. Pure thoracolumbar facet dislocation: Clinical features and CT appearance. Radiology 161:505, 1986. 46. Giacobetti, F Vaccaro, A.R.; Bos-Giacobetti, M.A.; et al. Vertebral .B.; artery occlusion associated with cervical spine trauma. A prospective analysis. Spine 22:188, 1997. 47. Goldberg, A.L.; Keaton, N.L.; Rothfus, W.E.; Daffner, R.H. Ankylosing spondylitis complicated by trauma: MR ndings correlated with plain radiographs and CT. Skeletal Radiol 22:333, 1996. 48. Gonzales, R.P.; Fried, P.O.; Bukhalo, M.; et al. Role of clinical examination in screening for blunt cervical spine injury. J Am Coll Surg 189:152, 1999. 49. Graber, M.A.; Kathol, M. Cervical spine radiographs in the trauma patient. Am Fam Physician 59:331, 1999. 50. Grant, G.A.; Mirza, S.K.; Chapman, J.R.; et al. Risk of early closed reduction in cervical spine subluxation injuries. J Neurosurg 90:13, 1999.
743
51. Grossman, M.D.; Reilly, P.M.; Gillet, T.; Gillett, D. National survey of the incidence of cervical spine injury and approach to cervical spine clearance in U.S. trauma centers. J Trauma 47:684, 1999. 52. Gupta, K.J.; Clancy, M. Discontinuation of cervical spine immobilization in unconscious patients with trauma in the intensive care unitsTelephone survey of practice in the south and west region. BMJ 314:1652, 1997. 53. Hall, A.J.; Wagle, V.G.; Raycroft, J.; et al. Magnetic resonance imaging in cervical spine trauma. J Trauma 34:21, 1993. 54. Handel, S.F Lee, Y.Y. Computed tomography of spinal fractures. .; Radiol Clin North Am 19:69, 1981. 55. Hanson, J.A.; Blackmore, C.C.; Mann, F .A.; Wilson, A.J. Cervical spine injury: Accuracy of helical CT used as a screening technique. Emerg Radiol 7:31, 2000. 56. Harris, J.H., Jr. Abnormal cervicocranial retropharyngeal soft-tissue contour in the detection of subtle acute cervicocranial injuries. Emerg Radiol 1:15, 1994. 57. Harris, J.H. The normal cervical spine. In: Harris, J.H., Jr.; Mirvis, S.E., eds. Radiology of Acute Cervical Spine Trauma, 3rd ed. Baltimore, Williams & Wilkins, 1995, p. 1. 58. Harris, J.H., Jr. Radiographic evaluation of spinal trauma. Orthop Clin North Am 17:75, 1986. 59. Harris, J.H., Jr.; Burke, J.T.; Ray, R.D.; et al. Low (type III) odontoid fracture: A new radiologic sign. Radiology 153:353, 1984. 60. Harris, J.H., Jr.; Carson, G.C.; Wagner, L.K. Radiologic diagnosis of traumatic occipitovertebral dissociation. 1. Normal occipitovertebral relationships on lateral radiographs of supine subjects. AJR 162:881, 1994. 61. Harris, J.H., Jr.; Carson, G.C.; Wagner, L.K.; Kerr, N. Radiologic diagnosis of traumatic occipitovertebral dissociation. 2. Comparison of three methods of detecting occipitovertebral relationships on lateral radiographs of supine subjects. AJR 162:887, 1994. 62. Harris, J.H. The radiographic examination. In: Harris, J.H.; Mirvis, S.E., eds. Radiology of Acute Cervical Spine Trauma, 3rd ed. Baltimore, Williams & Wilkins, 1995, p. 180. 63. Harris, J.H., Jr.; Harris, W.H.; Novelline, R.A., eds. The Radiology of Emergency Medicine. Spine, Including Soft Tissues of the Pharynx and Neck. Baltimore, Williams & Wilkins, 1993, p. 127. 64. Herman, G.T.; Liu, H.K. Display of three-dimensional information in computed tomography. J Comput Assist Tomogr 1:155, 1977. 65. Herr, C.H.; Ball, P.A.; Sargent, S.K.; Quinton, H.B. Sensitivity of prevertebral soft tissue measurement of C3 for detection of cervical spine fractures and dislocations. Am J Emerg Med 16:346, 1998. 66. Hildingsson, C.; Hietala, S.O.; Tollman, G. Scintigraphic ndings in acute whiplash injury of the cervical spine. Injury 20:265, 1989. 67. Hoffman, J.R.; Mower, W.R.; Wolfson, A.B.; et al. Validity of a set of clinical criteria to rule out injury to the cervical spine in patients with blunt trauma. National Emergency X-Radiology Utilization Study Group. N Engl J Med 343:94, 2000. 68. Holliman, C.J.; Mayer, J.S.; Cook, R.T.; et al. Is the anteroposterior cervical radiograph necessary in the initial trauma screening? AJR 9:421, 1991. 69. Iannacone, W.M.; DeLong, W.G.; Born, C.T.; et al. Dynamic computerized tomography of the occiput-atlas-axis complex in trauma patients with odontoid lateral mass asymmetry. J Trauma 3:1501, 1990. 70. Jabre, A. Subintimal dissection of the vertebral artery in subluxation of the cervical spine. Neurosurgery 29:912, 1991. 71. Jenkins, M.G.; Curran, P.; Rocke, L.G. Where do we go after the three standard cervical spine views in the conscious trauma patient? A survey. Eur J Emerg Med 6:215, 1999. 72. Kaneriya, P.P.; Schweitzer, M.E.; Spettell, C.; et al. The costeffectiveness of oblique radiography in the exclusion of C7-T1 injury in trauma patients. Skeletal Radiol 28:271, 1999. 73. Kassel, E.E.; Cooper, P.W.; Rubenstein, J.D. Radiology of spinal traumaPractical experience in a trauma unit. J Can Assoc Radiol 34:189, 1983. 74. Katz, M.A.; Beredjiklian, P.K.; Vresilovic, E.J.; et al. Computed tomographic scanning of cervical spine fractures: Does it inuence treatment? J Orthop Trauma 13:338, 1999. 75. Katz, R.C.; Wilson, L.; Fraser, N. Anxiety and its determinants in patients undergoing magnetic resonance imaging. J Behav Ther Exp Psychiatry 25:131, 1994.
744
SECTION II Spine 103. Modic, M.T.; Masaryk, T.J.; Ross, J.S. Magnetic Resonance Imaging of the Spine. Chicago, Year Book Medical, 1989. 104. Morris, R.E.; Hasso, A.N.; Thompson, J.R.; et al. Traumatic dural tears: CT diagnosis using metrizamide. Radiology 152:443, 1984. 105. Murray, J.B.; Ziervogel, M. The value of computed tomography in the diagnosis of atlantoaxial rotatory xation. Br J Radiol 63:894, 1990. 106. Nunez, D.B.; Ahmad, A.A.; Coin, C.G.; et al. Clearing the cervical spine in multiple trauma victims: A time-effective protocol using helical computed tomography. Emerg Radiol 1:273, 1994. 107. Nunez, D., Jr. Value of complete helical computed tomography scanning in identifying cervical spine injury in the unevaluable blunt trauma patients with multiple injuries. J Trauma 48;988, 2000. 108. Olerud, C.; Frost, A.; Bring, J. Spinal fractures in patients with ankylosing spondylitis. Eur Spine J 5:51, 1996. 109. Olshaker, J.S.; Barish, R.A. Acute traumatic cervical epidural hematoma. Ann Emerg Med 20:662, 1991. 110. Paleologos, T.S.; Fratzoglou, M.M.; Papadopoulos, S.S.; et al. Posttraumatic spinal cord lesions without skeletal or discal and ligamentous abnormalities: The role of MR imaging. J Spinal Disord 11:346, 1998. 111. Paley, D.; Schwartz, M.; Cooper, P.; et al. Fractures of the spine in diffuse skeletal hyperostosis. Clin Orthop 267:22, 1991. 112. Parent, A.D.; Harkey, H.L.; Touchstone, D.A.; et al. Lateral cervical spine dislocation and vertebral artery injury. Neurosurgery 31:501, 1992. 113. Pathria, M.N.; Petersilge, C.A. Spinal trauma. Radiol Clin North Am 29:847, 1991. 114. Pomeranz, S.J. Craniospinal Magnetic Resonance Imaging. Philadelphia, W.B. Saunders, 1989. 115. Post, M.J.; Green, B.A. The use of computed tomography in spinal trauma. Radiol Clin North Am 21:327, 1983. 116. Post, M.J.; Green, B.A.; Quencer, R.M.; et al. The value of computed tomography in spinal trauma. Spine 7:417, 1982. 117. Powers, B.; Miller, M.D.; Kramer, R.S.; et al. Traumatic anterior occipitoatlantal dislocation. Neurosurgery 4:127, 1979. 118. Pratt, E.S.; Green, D.A.; Spengler, D.M. Herniated intervertebral discs associated with unstable spine injuries. Spine 15:662, 1990. 119. Price, R.F Sellar, R.M.; Leung, C.; OSullivan, M.J. Traumatic .; vertebral arterial dissection and vertebrobasilar arterial thrombosis successfully treated with endovascular thrombolysis and stenting. AJNR 19:1677, 1998. 120. Quencer, R.M.; Bunge, R.P.; Egnor, M.; et al. Acute traumatic central cord syndrome: MRI-pathologic correlation. Neuroradiology 34:85, 1992. 121. Quencer, R.M.; Sheldon, J.J.; Post, M.J.D.; et al. MRI of the chronically injured cervical spinal cord. AJR 147:125, 1986. 122. Regenbogen, V.S.; Rogers, L.F Atlas, S.W.; et al. Cervical spinal .; cord injuries in patients with cervical spondylosis. AJR 146:277, 1986. 123. Reid, J.D.S.; Weight, J.A. Forty-three cases of vertebral artery trauma. J Trauma 28:1007, 1988. 124. Rizzolo, S.J.; Piazza, M.R.; Cotler, J.M.; et al. Intervertebral disc complicating cervical spine trauma. Spine 16:187, 1991. 125. Rizzolo, S.J.; Vaccaro, A.R.; Cotler, J.M. Cervical spine trauma. Spine 19:2288, 1994. 126. Ross, S.E.; Schwab, C.W.; Eriberto, T.D.; et al. Clearing the cervical spine. J Trauma 27:1055, 1987. 127. Roth, B.J.; Martin, R.R.; Foley, K.; et al. Roentgenographic evaluation of the cervical spine: A selective approach. Arch Surg 129:643, 1994. 128. Scarrow, A.M.; Levy, E.L.; Resnick, D.K.; et al. Cervical spine evaluation in obtunded or comatose pediatric trauma patients. A pilot study. Pediatr Neurosurg 30:169, 1999. 129. Schaeffer, D.M.; Flanders, A.E.; Osterholm, J.L.; et al. Prognostic signicance of magnetic resonance imaging in the acute phase of cervical spine injury. J Neurosurg 76:218, 1992. 130. Schwarz, N.; Buchinger, W.; Gaudernak, T.; et al. Injuries to the cervical spine causing vertebral artery trauma: Case reports. J Trauma 31:127, 1991. 131. Sees, D.W.; Rodriguez Cruz, L.R.; Flaherty, S.F Ciceri, D.P. The use .; of bedside uoroscopy to evaluate the cervical spine in obtunded trauma patients, J Trauma 45:768, 1998.
76. Katzberg, R.W.; Benedetti, P.F Drake, C.M.; et al. Acute cervical .; spine injuries: Prospective MR imaging assessment at a level 1 trauma center. Radiology 213:203, 1999. 77. Kaufman, H.H.; Harris, J.H., Jr.; Spencer, J.A.; et al. Metrizamideenhanced computed tomography and newer techniques of myelography. In: Bailey, R.W.; Sherk, H.H.; Dunn, E.J.; et al., eds. The Cervical Spine. Philadelphia, J.B. Lippincott, 1983, p. 103. 78. Kerslake, R.W.; Jaspan, T.; Worthington, B.S. Magnetic resonance imaging of spinal trauma. Br J Radiol 64:386, 1991. 79. Klein, G.R.; Vaccaro, A.R.; Albert, T.J.; et al. Efcacy of magnetic resonance imaging in the evaluation of posterior cervical spine fractures. Spine 24:771, 1999. 80. Koyanagi, I.; Iwasaki, Y.; Hida, K.; et al. Acute cervical cord injury without fracture or dislocation of the spinal column. J Neurosurg 93:15, 2000. 81. Kulkarni, M.V.; McArdle, C.B.; Kopanicky, D.; et al. Acute spinal cord injury: MR imaging at 1.5 T. Radiology 164:837, 1987. 82. Lee, C.; Woodring, J.H.; Rogers, L.F et al. The radiographic .; distinction of degenerative (spondylolisthesis and retrolisthesis) from traumatic slippage of the cervical spine. Skeletal Radiol 15:439, 1986. 83. Leidner, B.; Adeils, M.; Aspeln, P.; et al. Standardized CT examination of the multitraumatized patient. Eur Radiol 8:1630, 1998. 84. Leo, J.S.; Bergeron, R.T.; Kricheff, I.I.; et al. Metrizamide myelography for cervical spinal cord injuries. Radiology 129:707, 1978. 85. Lewis, L.M.; Docherty, M.; Ruoff, B.E.; et al. Flexion-extension in the evaluation of cervical spine injuries. Ann Emerg Med 20:117, 1991. 86. Louw, J.A.; Mafoyane, N.A.; Neser, C.P. Occlusion of the vertebral artery in cervical spine dislocations. J Bone Joint Surg Br 72:679, 1990. 87. Lynch, D.; McManus, F Ennis, J.T. Computed tomography in .; spinal trauma. Clin Radiol 37:71, 1986. 88. MacDonald, R.L.; Schwartz, M.L.; Mirich, D.; et al. Diagnosis of cervical spine injury in motor vehicle crash victims. How many x-rays are enough? J Trauma 30:392, 1990. 89. Mace, S.E. Unstable occult cervical spine fracture. Ann Emerg Med 20:1373, 1992. 90. Mascalchi, M.; Pozzo, G.D.; Dini, C.; et al. Acute spinal trauma: Prognostic value of MRI appearances at 0.5T. Clin Radiol 48:100, 1993. 91. Matin, P. The appearance of bone scans following fractures, including immediate and long-term studies. J Nucl Med 20:1227, 1979. 92. McAfee, P.C.; Yuan, H.A.; Fredrickson, B.E.; et al. The value of computed tomography in thoracolumbar fractures. J Bone Joint Surg Am 65:461, 1983. 93. McArdle, C.B.; Nicholas, D.A.; Richardson, C.J.; et al. Monitoring of the neonate undergoing MR imaging: Technical considerations. Radiology 159:223, 1986. 94. McArdle, C.B.; Wright, J.W.; Prevost, W.J.; et al. MR imaging of the acutely injured patient with cervical traction. Radiology 159:273, 1986. 95. McNamara, R.M.; Heine, E.; Esposito, B. Cervical spine injury and radiography in alert, high-risk patients. J Emerg Med 8:177, 1990. 96. Mirvis, S.E.; Borg, U.; Belzberg, H. MRI of ventilator-dependent patients: Preliminary experience. AJR 149:845, 1987. 97. Mirvis, S.E.; Diaconis, J.N.; Chirico, P.A.; et al. Protocol driven radiologic evaluation of suspected cervical spine injury: Efcacy study. Radiology 170:831, 1989. 98. Mirvis, S.E.; Geisler, F .H. Intraoperative sonography of cervical spinal cord injury. Results in 30 patients. AJNR 11:755, 1990. 99. Mirvis, S.E.; Geisler, F .H.; Jelinek, J.J.; et al. Acute cervical spine trauma: Evaluation with 1.5 T MR imaging. Radiology 166:807, 1988. 100. Mirvis, S.E.; Geisler, F .H.; Joslyn, J.N.; et al. Use of titanium wire in cervical spine xation as a means to reduce MR artifacts. AJNR 9:1229, 1988. 101. Mirvis, S.E.; Ness-Aiver, M. Magnetic resonance imaging of acute cervical spine trauma. In: Harris, J.H., Jr.; Mirvis, S.E., eds. Radiology of Acute Cervical Spine Trauma, 3rd ed. Baltimore, Williams & Wilkins, 1995, p. 114. 102. Mirvis, S.E.; Wolf, A.L. MRI of acute cervical spine trauma. Appl Radiol 21:15, 1992.
CHAPTER 26 Spinal Imaging 132. Seybold, E.A.; Dunn, E.J.; Jenis, L.G.; Sweeney, C.A. Variation on the posterior vertebral contour line at the level of C-2 on lateral cervical roentgenograms: A method for odontoid fracture detection. Am J Orthop 28:696, 1999. 133. Shaffer, M.A.; Doris, P.E. Limitation of the cross table lateral view in detecting cervical spine injuries: A retrospective analysis. Ann Emerg Med 10:508, 1981. 134. Shanmuganathan, K.; Mirvis, S.E.; Levine, A.M. Isolated articular pillar fractures of the cervical spine: Imaging observations in 20 patients. AJR 166:897, 1996. 135. Shanmuganathan, K.; Mirvis, S.E.; Levine, A.M. Rotational injury of the cervical facets: CT analysis of fracture patterns with implications for management and neurologic outcome. AJR 163:1156, 1994. 136. Shellock, F .G.; LipcZak, H.; Kamel, E. Monitoring patients during MR procedures. Appl Radiol 24:11, 1995. 137. Shellock, F .G.; Morisoli, S.; Kamel, E. MR procedures and biomedical implants, material, and devices. 1993 update. Radiology 189:587, 1993. 138. Silberstein, M.; Tress, B.M.; Henessey, O. Prevertebral swelling in cervical spine injury: Identication of ligament injury imaging. Clin Radiol 46:318, 1992. 139. Spain, D.A.; Trooskin, S.Z.; Flancbaum, L.; et al. The adequacy and cost-effectiveness of routine resuscitation area cervical spine radiographs. Ann Emerg Med 19:276, 1990. 140. Spitz, J.; Laer, I.; Tillet, K.; et al. Scintimetric evaluation of remodeling after fractures in man. J Nucl Med 34:1403, 1993. 141. Streiter, M.L.; Chambers, A.A. Metrizamide examination of traumatic lumbar nerve root meningocele. Spine 9:77, 1984. 142. Streitweisser, D.R.; Knopp, R.; Wales, L.R.; et al. Accuracy of standard radiographic views in detecting cervical spine fractures. Ann Emerg Med 12:538, 1983. 143. Takahashi, M.; Yamashita, Y.; Sakamoto, Y.; et al. Chronic cervical cord compression: Clinical signicance of increased signal on MRIs. Radiology 173:219, 1989. 144. Tarr, R.W.; Drolshagen, L.F Kerner, T.C. MR imaging of recent .; spinal trauma. J Comput Assist Tomogr 11:412, 1987. 145. Taylor, A.R.; Blackwood, W. Paraplegia in hyperextension cervical injuries with normal radiographic appearance. J Bone Joint Surg Br 30:245, 1948.
745
146. Templeton, P.A.; Young, J.W.R.; Mirvis, S.E.; et al. The value of retropharyngeal soft tissue measurements in trauma of the adult cervical spine. Skeletal Radiol 16:98, 1987. 147. Thomeir, W.C.; Brown, D.C.; Mirvis, S.E. The tilted odontoid: A sign of subtle odontoid fracture. AJNR 11:605, 1990. 148. Vannier, W.; Marsh, J.L.; Warren, J.O. Three-dimensional CT construction images for craniofacial surgical planning and evaluation. Radiology 150:179, 1984. 149. Veras, L.M.; Pedraza-Gutierrez, S.; Castellanos, J.; et al. Vertebral artery occlusion after cervical spine trauma. Spine 25:1171, 2000. 150. Wales, L.R.; Knopp, R.H.; Morishima, M.S. Recommendations for evaluation of the acutely injured cervical spine: A clinical radiologic algorithm. Ann Emerg Med 9:422, 1980. 151. Warner, J.; Shanmuganathan, K.; Mirvis, S.E.; et al. Magnetic resonance imaging of ligamentous injury of the cervical spine. Emerg Radiol 3:9, 1996. 152. Weir, D.C. Roentgenographic signs of cervical injury. Clin Orthop 109:9, 1975. 153. Weirich, S.D.; Cotler, H.B.; Narayana, P.A.; et al. Histopathologic correlation of magnetic resonance image signal patterns in a spinal cord injury model. Spine 15:630, 1990. 154. Weller, S.J.; Rossitch, E., Jr.; Malek, A.M. Detection of vertebral artery injury after cervical spine trauma using magnetic resonance angiography. J Trauma 46:660, 1999. 155. West, O.C.; Anbari, M.M.; Pilgram, T.K.; Wilson, A.J. Acute cervical spine trauma: Diagnostic performance of single-view versus three-view radiographic screening. Radiology 204:819, 1997. 156. Willis, B.K.; Greiner, F Orrison, W.W.; et al. The incidence of .; vertebral artery injury after midcervical spine fracture or subluxation. Neurosurgery 34:435, 1994. 157. Yeakley, J.; Edeiken-Monroe, B.; Harris, J.H., Jr. Computed tomography of spinal trauma and degenerative disease. Instr Course Lect 34:85, 1985. 158. Young, J.W.R.; Resnick, C.S.; DeCandido, P.; et al. The laminar space in the diagnosis of rotational exion injuries of the cervical spine. AJR 152:103, 1989. 159. Zabel, D.D.; Tinkoff, G.; Wittenborn, W.; et al. Adequacy and efcacy of lateral cervical radiography in alert, high-risk blunt trauma patient. J Trauma 43:952, 1997.

CH 26

Hochgeladen von

Dokumentinformationen

Originalbeschreibung:

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

CH 26

Hochgeladen von

Copyright:

Verfügbare Formate

C H A P T E R zzzzzzzzzzzzzzzzzzzzzzzzzzz

DIAGNOSTIC IMAGING OF THE CERVICAL SPINE

Imaging Modalities: An Overview

Copyright 2003 Elsevier Science (USA). All rights reserved.

CHAPTER 26 Spinal Imaging

CLINICAL ASSESSMENT RESULT

Normal examination No cervical spine pain Alert No major distracting injuries

Neurologic deficit consistent with cervical cord/root injury

Neurologic examination normal Neck pain or tenderness to motion or palpation

Supine AP, OMO, lateral cervical radiographs1

Unreliable clinical assessment; obtunded, unconscious, major distracting injuries

Supine AP, OMO, lateral cervical radiographs2

Supine AP, lateral cervical radiographs3 Print Graphic

No imaging assessment recommended

Maintain cervical immobilization, obtain consultation to plan further evaluation

Physician supervised flexion-extension lateral radiographs

Total cervical spiral CT

Normal Normal Cervical MRI

Repeat flexion-extension views in 710 days if symptoms persist

Plain Film Radiography of the Cervical Spine

Copyright 2003 Elsevier Science (USA). All rights reserved.

Copyright 2003 Elsevier Science (USA). All rights reserved.

CHAPTER 26 Spinal Imaging

Copyright 2003 Elsevier Science (USA). All rights reserved.

Copyright 2003 Elsevier Science (USA). All rights reserved.

CHAPTER 26 Spinal Imaging

Copyright 2003 Elsevier Science (USA). All rights reserved.

Print Graphic Print Graphic

FLEXION-EXTENSION STRESS VIEWS

Copyright 2003 Elsevier Science (USA). All rights reserved.

CHAPTER 26 Spinal Imaging

Copyright 2003 Elsevier Science (USA). All rights reserved.

Copyright 2003 Elsevier Science (USA). All rights reserved.

CHAPTER 26 Spinal Imaging

IMAGING APPROACH TO THE POTENTIALLY INJURED CERVICAL SPINE

Spinal Imaging of the Polytraumatized Patient: An Overview

Copyright 2003 Elsevier Science (USA). All rights reserved.

Copyright 2003 Elsevier Science (USA). All rights reserved.

CHAPTER 26 Spinal Imaging

Copyright 2003 Elsevier Science (USA). All rights reserved.

Copyright 2003 Elsevier Science (USA). All rights reserved.

CHAPTER 26 Spinal Imaging

IMAGING ASSESSMENT OF POTENTIAL INJURY TO THE NONCERVICAL SPINE

Plain Film Radiography of the Thoracic Spine

Copyright 2003 Elsevier Science (USA). All rights reserved.

Plain Film Radiography of the Lumbar Spine

Copyright 2003 Elsevier Science (USA). All rights reserved.

CHAPTER 26 Spinal Imaging

Plain Radiography of the Sacrum and Coccyx

DIAGNOSTIC MODALITIES IN IMAGING SPINAL TRAUMA: ADVANTAGES AND LIMITATIONS

Computed Tomography of the Spine

Copyright 2003 Elsevier Science (USA). All rights reserved.

Computed Tomography with Intrathecal Contrast

Copyright 2003 Elsevier Science (USA). All rights reserved.

CHAPTER 26 Spinal Imaging

Computed Tomography with Two-Dimensional Multiplanar and Three-Dimensional Re-formation

Copyright 2003 Elsevier Science (USA). All rights reserved.

Magnetic Resonance Imaging of the Spine

Copyright 2003 Elsevier Science (USA). All rights reserved.

CHAPTER 26 Spinal Imaging

Copyright 2003 Elsevier Science (USA). All rights reserved.

Copyright 2003 Elsevier Science (USA). All rights reserved.