Published online before print July 1, 2005, 10.1148/radiol.2362040281
(Radiology 2005;236:601-608.)
© RSNA, 2005
Chance-Type Flexion-Distraction Injuries in the Thoracolumbar Spine: MR Imaging Characteristics1
Clare J. Groves, FRCR,
Victor N. Cassar-Pullicino, FRCR,
Bernhard J. Tins, FRCR,
Prudencia N. M. Tyrrell, FRCR and
Iain W. McCall, FRCR
1 From the Department of Radiology, Robert Jones and Agnes Hunt Orthopaedic and District Hospital, Oswestry, Shropshire SY10 7AG, England. Received February 13, 2004; revision requested April 20; revision received September 22; accepted January 17, 2005.
Address correspondence to V.N.C.P. (Victor.Pullicino{at}rjah.nhs.uk).
 |
ABSTRACT
|
|---|
PURPOSE: To evaluate retrospectively the magnetic resonance (MR) imaging features of Chance-type flexion-distraction injuries.
MATERIALS AND METHODS: The authors' institutional review board does not require its approval or patient informed consent for retrospective studies. Imaging data were reviewed retrospectively for 24 patients (15 male, nine female; mean age, 28 years; range, 9–71 years) who had sustained radiographically typical Chance-type flexion-distraction injuries. The posterior vertebral body height remained unchanged or was increased in these patients. Two radiologists recorded a variety of bone and soft-tissue abnormalities seen with MR imaging. Based on consensus, the documented findings were sequentially analyzed to determine their frequencies.
RESULTS: Combined bone and soft-tissue injuries occurred in 23 (96%) of 24 patients, were more common than soft-tissue damage alone (one [4%] of 24 patients), and occurred primarily at the thoracolumbar junction. Contiguous vertebral injury was seen in 20 (83%) of 24 patients, usually in the form of anterosuperior vertebral endplate edema, while noncontiguous injury occurred in eight (33%) of 24 patients. Extensive subcutaneous and paraspinal muscle edema was seen in all patients and extended over several segments. Posterior osteoligamentous complex disruption also occurred in all patients. Horizontally oriented fractures of the posterior neural arches produced an MR imaging pattern that the authors call the sandwich sign, which consists of linear hemorrhage framed by marrow edema. This sign was seen in 12 (50%) of 24 patients. In seven (29%) of 24 patients, a fracture line extending from a damaged pedicle was seen to exit through the contralateral posterosuperior aspect of the vertebral body, with extension of the fracture fragments into the spinal canal.
CONCLUSION: A spectrum of features is discernible with MR imaging in Chance-type injuries.
© RSNA, 2005
|
INTRODUCTION
|
|---|
Chance-type spinal injuries result from hyperflexion of the spine around an anterior fulcrum in combination with a posterior vertical distraction force. Typically, the spinal column splits transversely in the posteroanterior direction. The split may involve the bone components of the spine or the diskoligamentous structures, or there may be a combination of bone and soft-tissue disruption (Fig 1). The radiographic appearances of this injury (Fig 2) are well documented (1–3), with scarce reference to the magnetic resonance (MR) imaging features.
View larger version (32K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1a. Variations of Chance-type flexion-distraction spinal injuries include (a) classic Chance fracture, (b) fulcrum fracture,and (c) pure soft-tissue flexion-distraction injury.
|
|
View larger version (33K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1b. Variations of Chance-type flexion-distraction spinal injuries include (a) classic Chance fracture, (b) fulcrum fracture,and (c) pure soft-tissue flexion-distraction injury.
|
|
View larger version (32K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1c. Variations of Chance-type flexion-distraction spinal injuries include (a) classic Chance fracture, (b) fulcrum fracture,and (c) pure soft-tissue flexion-distraction injury.
|
|
View larger version (150K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2. Classic Chance fracture. Lateral radiograph shows distraction fracture that extends horizontally through the pedicle (right arrow) and into the vertebral body. The fractured vertebral body has not been fully reduced, and the height of its posterior aspect is increased. There is an anterosuperior wedge-shaped compression fracture (left arrow) in the same vertebral body.
|
|
Thoracolumbar spinal injuries are classified by means of radiography and computed tomography (CT) and without the use of MR imaging, primarily on the basis of the presumed mechanical mode of failure of the vertebral bodies (4). These classification systems have been shown to be unreliable when correlated with MR imaging findings, surgical findings, and surgical outcomes (5–7). The multiplicity of vectors involved in these complex injuries renders a mechanistic classification terminology confusing and misleading.
At our tertiary referral center for acute spinal injury, we have been aware, since the advent of MR imaging, of classification difficulties with spinal injuries referred to our institution. We have observed similarities in the vertebral body injury pat-terns caused by flexion-distraction forces (Chance-type fractures) and those produced by axial loading (burst fractures). The distinction between the two types of spinal injury is important, since Chance-type injuries carry a higher risk of associated visceral damage, spinal cord trauma, and concomitant spinal injury (2,8–12). Thus, the label applied to the injury by the radiologist may imply a different treatment algorithm to the clinician.
The role of MR imaging in differentiating flexion-distraction injuries from pure burst fractures may be crucial. The purpose of our study was to evaluate retrospectively the MR imaging features of Chance-type flexion-distraction injuries.
|
MATERIALS AND METHODS
|
|---|
Patients
In this retrospective study, patients were selected from hospital and medical databases for the period 1997–2003. Our institutional review board does not require its approval or patient informed consent for retrospective studies such as this one. The clinical details for 30 patients in whom Chance injury or flexion-distraction injury had been diagnosed were obtained from the hospital's electronic patient record system. For each patient, radiographs, CT scans, and MR images were available for review by two authors (C.J.G., fellow in musculoskeletal imaging, and V.N.C.P., consultant radiologist, who had 5 and more than 15 years of experience in MR imaging of the spine, respectively). They reviewed the images to ensure that recognized radiographic criteria of Chance-type injuries were present.
Smith and Kaufer (13) stated that Chance-type injuries are characterized by the following: (a) disruption of the posterior elements of the spine, which may be osseous, ligamentous, or both; (b) longitudinal separation of the disrupted posterior elements; (c) minimal or no decrease in the anterior vertical height of the involved vertebral body; (d) minimal or no forward displacement of the superior vertebral fragment or vertebra; and (e) minimal or no lateral displacement of this fragment or the superior vertebra. In addition to the criteria of Smith and Kaufer, we included one based on an observation by Ferguson and Allen (4) that the height of the posterior aspect of the affected vertebral body is equal to or greater than that of the adjacent inferior vertebral body. On the basis of these criteria, six patients originally classified as having Chance-type injuries were removed from the study because of vertebral body fractures in which the posterior vertebral body height was reduced. The remaining 24 patients included 15 male and nine female patients with a mean age of 28 years (range, 9–71 years).
MR Imaging
Three of the 24 patients in the study underwent MR imaging examinations at other institutions, which did not cover the cervical spine. Full technical data were not available for these patients, but the sequences and section thicknesses employed were similar to those used at our institution. The remaining 21 patients underwent whole-spine imaging performed with a 1.5-T MR imager (Magnetom Vision; Siemens Medical Systems, Erlangen, Germany) and with standard thoracolumbar protocols. Sagittal images were acquired with a 4-mm section thickness, a 380-mm field of view, a 258 x 512 matrix, and the following sequences: T1-weighted turbo spin echo (SE) (600/12 [repetition time msec/echo time msec]), T2-weighted turbo SE (5000/120), and short inversion time inversion-recovery (STIR) (2915/30/150 [repetition time msec/echo time msec/inversion time msec]). Transverse images were acquired with a 4-mm section thickness, a 192 x 256 matrix, and the following sequences: T1-weighted SE (750/15) and T2-weighted turbo SE (5500/120). The cervical and thoracic spine from C1 through T8 was imaged with 3-mm sections and similar sagittal presets for T1-weighted, T2-weighted, and STIR sequences, although the matrix was reduced to 256 x 256 for the STIR sequence. Transverse images from C1 through T8 were acquired with a field of view of 250 mm and the following sequences: T1-weighted SE (750/15) and T2-weighted gradient echo (703/15; 15° flip angle).
MR Image Review
There was an average interval of 4 days (range, 1–14 days) between the initial injury and MR imaging. The MR images of the 24 patients were reviewed, first independently and then in consensus, by the same two authors who reviewed the radiographs and CT images (C.J.G., V.N.C.P.). Data were recorded on a pro forma spreadsheet on which the soft-tissue and osseous signs of spinal injury were listed. The interval from injury to imaging was recorded, along with the levels of injury for each patient. Particular note was made of the presence and levels of contiguous and noncontiguous injuries. The craniocaudal extent of edema in the subcutaneous fat planes, presence or absence of paravertebral fluid collection, presence or absence of edema of the paravertebral musculature, and signs of ligamentous damage (discontinuity, separation, edema, or hemorrhage) (14) were also noted.
The posterior osteoligamentous complex consists of the supraspinous ligament, the interspinous ligament, the flaval ligaments, the capsules of the facets, and the posterior neural bone arch, which includes the pedicles (9,11). Disks were assessed for normal morphologic features or anatomic disruption. The spinal cord was assessed for normal morphologic features and the presence of hemorrhage or edema, and the presence or absence of epidural hemorrhage was noted. Fractures of the lamina and spinous processes were recorded. Facet joints were assessed for normal morphologic features or anatomic disruption, while the pedicles were assessed for normal morphologic features, edema, hemorrhage, and a central area of low signal intensity encased by linear high signal intensity (sandwich sign) on T2-weighted images and STIR images. Vertebral body edema was recorded according to its distribution and the proportion of the vertebral body involved, while vertebral body fracture patterns were recorded descriptively. The MR sequence that best demonstrated each category of injury was noted for all 24 patients.
Data Analysis
This study was designed as a descriptive study, and statistical analysis was not performed. After the reviewers reached consensus, the results were expressed as percentage frequencies. No attempt was made to measure inter- or intraobserver variability.
|
RESULTS
|
|---|
In three of the 24 patients with Chance-type spinal injuries, the cervical and proximal thoracic vertebrae (C1 through T3) were not included in the MR imaging examination. In the remaining 21 patients, MR imaging of the whole spine had been performed. T1-weighted, T2-weighted, and STIR images were available for all patients in the sagittal and transverse planes. The vertebral body and displaced posterior neural arch fractures were best appreciated on the T1-weighted images, whereas pedicular injury and soft-tissue disruption were best seen on T2-weighted and STIR images. The most common level of injury was between T12 and L2, and only four patients had sustained injuries between T6 and T9. One patient had sustained purely soft-tissue flexion-distraction injuries. One patient had Chance fractures at two levels (L1 and L2), one patient had a Chance fracture at L1 with a burst fracture at T12, and in one patient with an immature spine, the inferior endplate had distracted from the main vertebral body. The remaining 19 patients had incurred classic bone Chance-type injuries at a single level, all of whom exhibited an increase in the posterior vertebral height of the fractured vertebral body.
Injuries to the Posterior Osteoligamentous Complex
On the MR images of every patient, the supraspinous ligament was ruptured, there was extensive edema of the interspinous ligaments, and the ligamentum flavum was disrupted (Fig 3). Only one patient retained intact pedicles, as he had sustained a soft-tissue Chance-type injury with disruption of the T12-L1 disk and no bone fracture.
View larger version (156K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a. Soft-tissue injuries. (a) Transverse T2-weighted turbo SE MR image (5000/120) demonstrates extensive paraspinal muscle edema, psoas muscle edema (upper white arrow), and erector spinae muscle edema (lower white arrow). An intraspinal hematoma (black arrow) has displaced the conus anteriorly. (b) Sagittal T2-weighted turbo SE image (5000/120) shows posterior ligamentous complex disruption, with torn supraspinous ligament (small white arrow), interspinous ligament (small black arrow), and ligamentum flavum (large black arrow). Note the mild disruption in the anterior part of the L1-2 disk and buckling of the anterior longitudinal ligament (large white arrow). (c) Sagittal STIR image (2915/30/150, 180° flip angle, 350-mm field of view) shows extensive disruption of subcutaneous fat plane (black arrow) in association with torn interspinous ligaments at T12-L1 (white arrow). Note the intact diskovertebral morphologic structure.
|
|
View larger version (98K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b. Soft-tissue injuries. (a) Transverse T2-weighted turbo SE MR image (5000/120) demonstrates extensive paraspinal muscle edema, psoas muscle edema (upper white arrow), and erector spinae muscle edema (lower white arrow). An intraspinal hematoma (black arrow) has displaced the conus anteriorly. (b) Sagittal T2-weighted turbo SE image (5000/120) shows posterior ligamentous complex disruption, with torn supraspinous ligament (small white arrow), interspinous ligament (small black arrow), and ligamentum flavum (large black arrow). Note the mild disruption in the anterior part of the L1-2 disk and buckling of the anterior longitudinal ligament (large white arrow). (c) Sagittal STIR image (2915/30/150, 180° flip angle, 350-mm field of view) shows extensive disruption of subcutaneous fat plane (black arrow) in association with torn interspinous ligaments at T12-L1 (white arrow). Note the intact diskovertebral morphologic structure.
|
|
View larger version (99K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3c. Soft-tissue injuries. (a) Transverse T2-weighted turbo SE MR image (5000/120) demonstrates extensive paraspinal muscle edema, psoas muscle edema (upper white arrow), and erector spinae muscle edema (lower white arrow). An intraspinal hematoma (black arrow) has displaced the conus anteriorly. (b) Sagittal T2-weighted turbo SE image (5000/120) shows posterior ligamentous complex disruption, with torn supraspinous ligament (small white arrow), interspinous ligament (small black arrow), and ligamentum flavum (large black arrow). Note the mild disruption in the anterior part of the L1-2 disk and buckling of the anterior longitudinal ligament (large white arrow). (c) Sagittal STIR image (2915/30/150, 180° flip angle, 350-mm field of view) shows extensive disruption of subcutaneous fat plane (black arrow) in association with torn interspinous ligaments at T12-L1 (white arrow). Note the intact diskovertebral morphologic structure.
|
|
The imaging features of pedicular injury varied from simple bone marrow edema to a combination of edema and hemorrhage that we termed the sandwich sign. The sandwich sign consisted of a linear or triangular layer of low signal intensity (hemorrhage) framed by two layers of high signal intensity (edema) on the T2-weighted images and STIR images (Fig 4). This sign was not seen on the T1-weighted images, except in two instances of subacute hemorrhage that appeared as an area of high signal intensity with areas of low signal intensity (edema) on either side on T1-weighted MR images obtained 10–14 days after injury. The sandwich sign was observed in both pedicles in seven patients and was unilateral in five patients in whom the contralateral pedicle was edematous but pedicular structure was preserved. The overall frequency of the sandwich sign was therefore 19 (40%) of 48 pedicles, and it was present in 12 (50%) of 24 patients.
View larger version (112K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a. MR imaging features of the sandwich sign. (a) Sagittal STIR MR image (2915/30/150) shows the sandwich sign (black arrow): an area of low-signal-intensity hemorrhage within the pedicular fracture, encased by a rim of edema. Note the linear areas of edema in the two adjacent injured vertebral bodies (white arrows). (b) Sagittal T2-weighted turbo SE image (5000/120) shows a horizontal fracture in the pedicle of T12, with linear adjacent zones of high and low signal intensity that produce a sandwich sign (arrow). The fracture extends into the vertebra and is associated with an anterosuperior wedge-shaped fracture of the vertebral body.
|
|
View larger version (113K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b. MR imaging features of the sandwich sign. (a) Sagittal STIR MR image (2915/30/150) shows the sandwich sign (black arrow): an area of low-signal-intensity hemorrhage within the pedicular fracture, encased by a rim of edema. Note the linear areas of edema in the two adjacent injured vertebral bodies (white arrows). (b) Sagittal T2-weighted turbo SE image (5000/120) shows a horizontal fracture in the pedicle of T12, with linear adjacent zones of high and low signal intensity that produce a sandwich sign (arrow). The fracture extends into the vertebra and is associated with an anterosuperior wedge-shaped fracture of the vertebral body.
|
|
Facet dislocation was common in our series, with 27 (56%) of 48 facet joints showing evidence of capsular discontinuity and an "empty facet" sign on the transverse images. In seven patients with unilateral facet disruption, the contralateral pedicle was fractured. Only five patients had bilateral intact facet joints, but all of them had bilateral pedicular damage. Most cases of cord injury (seven of eight) were seen in patients who had sustained either bilateral facet dislocation or bilateral pedicular injury.
Vertebral Body Injuries
In all cases, the injured vertebral body itself was morphologically abnormal at the level of the injury, with the abnormality involving predominantly the anterosuperior corner. There was a loss of anterior vertebral body height in 20 (83%) of 24 patients because of either compression or corner fracture. Edema of the marrow was evident, a finding that was in keeping with trabecular injury within the vertebral body, but the distribution of marrow edema (on T1-weighted and STIR images) was mainly linear and horizontally oriented and involved less than 50% of the vertebral body area in the craniocaudal plane (Fig 4). The degree of edema did not appear to be related to the interval between injury and MR imaging.
In seven patients in whom a single pedicle was fractured transversely, the fracture line extended into and across the vertebral body from the damaged pedicle. The fracture exited at the contralateral superior endplate, through the posterosuperior aspect of the vertebral body in five patients and more anteriorly along the superior endplate in the other two. These posteriorly located fractures were associated with bone displacement into the spinal canal, but the posterior vertebral body height was unchanged or increased when compared with the adjacent vertebrae (Fig 5). The disks adjacent to the fracture line were disrupted in five of the seven patients.
View larger version (103K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5. Posterior vertebral fracture fragment. Sagittal T2-weighted turbo SE MR image (5000/120) demonstrates a displaced posterosuperior fracture fragment (long white arrow) associated with an increase in the posterior vertebral body height at L2. Note also the L1-2 disk disruption (short white arrow) and extensive ligamentous disruption of the posterior osteoligamentous complex at L1-2 (arrowhead) with subcutaneous soft-tissue injury that extends over a large number of vertebral segments (black arrows).
|
|
Concomitant vertebral trauma was common, with contiguous injury to the adjacent vertebrae or disks on either side of the primary injury (Fig 6) demonstrated in 20 (83%) of 24 patients. The most consistent finding was edema, usually in the anterosuperior aspect of the vertebral bodies and endplates (Fig 4). Morphologically intact vertebral bodies indicated that this edema was due to bone bruising. One patient had a contiguous burst injury. Linear fracture of the adjacent vertebral bodies was seen in three patients, with contiguous spinous process fracture in another two. Contiguous laminar fracture or edema was seen in five patients.
View larger version (92K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6. Contiguous injury. Sagittal T1-weighted SE MR image (600/12) demonstrates horizontal fracture through the vertebral body of L2 in association with anterior vertebral body wedging and buckling of the anterior longitudinal ligament. Note the contiguous linear horizontally oriented fractures in the vertebral bodies of L1 and L3 (arrows) and injury to the posterior osteoligamentous complex (arrowhead).
|
|
Noncontiguous injuries were seen less commonly, in eight (33%) of 24 patients. This figure may be an underestimate, since the cervicothoracic spine was not imaged in three patients. However, the distribution of noncontiguous injury was wide. Flexion-distraction injury at L1 was accompanied by a wedge-shaped compression fracture of the vertebral body of L5 in one patient and an endplate compression fracture with edema at T4 and T5 in another patient. A horizontally oriented spinous process fracture at L2 was seen in one patient in association with a Chance-type fracture of T11. In another patient, a Chance-type fracture at L1 was seen in association with disruption of the posterior osteoligamentous complex at C6-7 and a fluid-fluid level representing hemorrhage into the posterior paraspinal soft tissues (Fig 7). The distribution of bone injuries is illustrated in Table 1.
View larger version (144K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7a. Noncontiguous cervical injury following Chance-type injury at L2. (a) Sagittal T2-weighted turbo SE MR image (5000/120) shows supraspinous ligament disruption at C7-T1 (short black arrow) with ligamentum flavum tear at C6-7 (white arrow). Note fluid-fluid level due to acute hematoma in cleft within the subcutaneous fat (long black arrow). (b) Sagittal T2-weighted turbo SE image (5000/120) shows extensive two-level disruption of the posterior osteoligamentous complex in association with a Chance-type injury at T11 (black arrows), cord injury, epidural hemorrhage within the spinal canal, and fracture through the vertebral body. Note the horizontal fracture through the spinous process of L2 (white arrow).
|
|
View larger version (101K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7b. Noncontiguous cervical injury following Chance-type injury at L2. (a) Sagittal T2-weighted turbo SE MR image (5000/120) shows supraspinous ligament disruption at C7-T1 (short black arrow) with ligamentum flavum tear at C6-7 (white arrow). Note fluid-fluid level due to acute hematoma in cleft within the subcutaneous fat (long black arrow). (b) Sagittal T2-weighted turbo SE image (5000/120) shows extensive two-level disruption of the posterior osteoligamentous complex in association with a Chance-type injury at T11 (black arrows), cord injury, epidural hemorrhage within the spinal canal, and fracture through the vertebral body. Note the horizontal fracture through the spinous process of L2 (white arrow).
|
|
Soft-Tissue Injuries
Soft-tissue signs of the flexion-distraction force included disruption with cleft separation of the attachment of the subcutaneous fat from the tips of the spinous processes over consecutive spinal levels (Table 2). In all cases, extensive subcutaneous disruption resulted in the MR signal intensity features of edema and hemorrhage, which filled the cleft within the fractured adipose tissue and involved six to 19 segments (Fig 3c). The degree of edema was not related to the interval between injury and imaging. The erector spinae muscles were edematous in 23 (96%) of 24 patients. Psoas muscle edema was present in one-third of patients, always in association with paraspinal muscle edema (Fig 3a), and both conditions were bilateral and frequently symmetric. Although paravertebral hematoma was found in 20 (83%) of 24 patients, it was characteristically small and took the form of a thin rim around the vertebral body. Epidural hematoma was common (Fig 8) and occurred anteriorly, posteriorly, or in both components of the epidural space. Areas of hematoma, with a characteristic elongated appearance, were present in 21 (88%) of 24 patients.
View larger version (115K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8. Disk disruption and epidural hematoma after flexion-distraction injury. Sagittal T2-weighted turbo SE MR image (5000/120) demonstrates extensive posterior ligamentous disruption in association with extensive extradural hemorrhage (long white arrow), posterior annular disruption of the T11-12 disk (long black arrow), and a fracture through the superior endplate of T12 (short white arrow). Note the high signal intensity within the cord at this level (short black arrow).
|
|
Seven patients had cord edema, and one had hemorrhage within the cord; no cauda equina damage was seen. The seven cases of cord damage included one pure soft-tissue Chance-type injury in which the injury dislocated both facet joints and exited through the T12-L1 disk with no bone involvement. The injured segment of cord was localized to the level of vertebral or disk injury in seven patients (Fig 8), while altered signal intensity in the cord in one patient extended over five segments proximal to the level of the vertebral injury.
In 13 (54%) of 24 patients, we had difficulty assessing the posterior longitudinal ligament and were unable to exclude posterior longitudinal ligament disruption confidently. Even in cases in which the widened intervertebral disk space and the separated fractured pedicular components indicated complete posterior longitudinal ligament rupture, we could not identify the free ends. In these circumstances, the elongated appearance of hemorrhage in the anterior epidural space and of venous engorgement may preclude clear visualization of the damaged ligament. The anterior longitudinal ligament remained intact in most patients (20 of 24) and could be seen to buckle rather than tear (Fig 3). However, complete anterior longitudinal ligament rupture occurred in four patients, all of whom sustained bilateral fractured pedicles and disk injury. Cord injury was documented in two of these four patients.
|
DISCUSSION
|
|---|
Chance-type spinal injuries are thought to result from hyperflexion of the spine around a fulcrum. They were originally described in association with high-speed road traffic accidents in which the victim had been restrained by means of a lap belt (3). The classic radiographic features of this injury include horizontal fractures of the pedicles. The fracture line has been shown to extend transversely through the vertebral body (13). In cases in which reduction of the fracture is incomplete, the posterior vertebral body height is increased in relation to the adjacent inferior vertebral body (4). The static or increased posterior vertebral body height that is found in Chance-type injury allows it to be distinguished from burst fracture, in which the posterior vertebral body height is reduced (4).
A small wedge-shaped compression fracture of the anterosuperior vertebral body has been described (3), which suggests that the fulcrum point is located behind the anterior column (15). It is intuitive that profound hyperflexion of the spine would stretch the overlying soft tissues and then distract and tear the spinal ligaments before splitting the osseous spinal column. However, injury is seldom caused by a single vector, and it is self-evident that rotational and deceleration forces occur in addition to the primary flexion-distraction insult. A spectrum of injury patterns must be expected, but the key indication that a flexion-distraction injury has occurred is posterior osteoligamentous complex disruption in the presence of minimal vertebral malalignment. A transverse orientation of fracture lines through the vertebral body or pedicles also is in keeping with this type of trauma, but it must be remembered that the line of disruption may pass through the capsule of the facet joint and into the disk without involving bone (2,13).
MR imaging allows depiction of the injured posterior osteoligamentous component that is the hallmark of Chance-type injuries. There is widespread damage to the subcutaneous tissues, as highlighted by edema and hemorrhage in the subcutaneous fat. This is extensive, is best seen in the acute phase, involves multiple contiguous vertebral levels, and is seen irrespective of the type of osseous and/or soft-tissue injury. The depiction of the torn supraspinous, interspinous, and flaval ligaments is equally well depicted in the acute stage and is especially striking when there is residual posterior interosseous separation. In instances in which spontaneous reduction has taken place at rest, the injury is readily identifiable in these tissues with MR imaging, even though radiographs appear normal. Edema in the paraspinal musculature is common and would be in keeping with stretching and tearing of the muscles during the hyperflexion phase of injury.
The superior depiction of cord status by MR imaging is undisputed, irrespective of the type of spinal trauma. Twenty-nine percent of our patients (seven of 24) sustained damage to the spinal cord at the level of the Chance-type injury. MR imaging is useful in showing the site and extent of spinal cord damage, as it can be located proximal to the level of osteoligamentous trauma. Other investigators reported a substantial frequency of neurologic deficit after fracture at the thoracolumbar junction, ranging from 30% to 40% (2,16). Most cases of cord injury in our series (seven of eight) were associated with either bilateral facet dislocation or bilateral pedicular fractures. Five of these patients also had disk injury, and the anterior longitudinal ligament was ruptured in two. These findings suggest that the cord damage occurs as a result of profound hyperflexion in which all three columns and the cord are placed under enormous distractive tension.
Of particular importance in our study was the high level of damage to adjacent vertebrae depicted by MR images, with contiguous injury demonstrated in 20 (83%) of 24 patients. The majority of contiguous injuries were due to compressive forces, with bone edema occurring in the anterosuperior vertebral body. MR images in eight (33%) of 24 patients demonstrated noncontiguous vertebral injury, a result that correlates well with the 39% frequency of noncontiguous injuries reported by Qaiyum et al (17). These findings of noncontiguous injuries throughout the spine emphasize the need to image the whole spine in any thoracolumbar injury.
In injuries where a horizontal fracture of the pedicle has occurred, the pedicle is widened in craniocaudal extent, and the resultant hemorrhage within the fracture line is shown with MR imaging. The majority of our patients were imaged about 4 days after injury, and at this stage the hemorrhage within the fracture demonstrated low signal intensity on T2-weighted images and STIR images. The area of hemorrhage was encased by a rim of high signal intensity in the adjacent fracture margins, a finding that was presumed to represent edema. The resultant sandwich sign was present in 12 (50%) of our 24 patients. The MR appearance of this sign is primarily governed by the choice of sequence parameters, the time interval between injury and MR imaging, and the size of the gap in the horizontally distracted and fractured pedicle. The high frequency of the sandwich sign, coupled with the absence of its description in other spinal injuries, makes this sign a useful MR imaging indication of flexion-distraction Chance-type fracture. In instances where the pedicular fracture extends into the posterior aspect of the vertebral body, an intravertebral sandwich sign may result. The sign also can be seen in horizontal fractures of the laminae.
Fracture patterns within the vertebral body itself were highly variable in our cohort of 24 patients. A pure horizontal fracture or trabecular fracture, shown by linear edema in the vertebral body, occurred in some patients. In other patients, the fracture line extended obliquely from a fractured pedicle upward through the vertebral body to exit at the endplate. Fracture fragments were seen to extend into the canal in some patients, and these injuries could easily be misinterpreted as burst fractures at CT or radiography. However, the increase in posterior vertebral body height and disruption of the posterior osteoligamentous complex in these patients confirmed that these fractures were part of the flexion-distraction injury spectrum. Given findings reported by previous authors (18,19), we believe that the primary hyperflexion injury that splits the vertebral body is followed by a high-velocity recoil injury, with fragmentation of the vertebral body occurring as a secondary event (Fig 9). Our initial decision to exclude six patients because of reduced posterior vertebral body height may have been incorrect, since their injuries may have been part of the wide spectrum of vertebral body injuries produced by flexion-distraction.
View larger version (35K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9a. Diagrams show morphologic differences in vertebral fractures with flexion-distraction injuries. Arrows indicate the direction of movement. (a) Low-velocity injury. Classic Chance fracture extends horizontally into the vertebral body. Low-velocity recoil reduces the fracture to near-anatomic position with no further bone disruption. (b) High-velocity injury. After the primary horizontal fracture of the vertebral body during flexion, high-velocity recoil brings the superior and inferior parts of the fractured vertebral body together again with great force, which results in a secondary burst fracture.
|
|
View larger version (36K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9b. Diagrams show morphologic differences in vertebral fractures with flexion-distraction injuries. Arrows indicate the direction of movement. (a) Low-velocity injury. Classic Chance fracture extends horizontally into the vertebral body. Low-velocity recoil reduces the fracture to near-anatomic position with no further bone disruption. (b) High-velocity injury. After the primary horizontal fracture of the vertebral body during flexion, high-velocity recoil brings the superior and inferior parts of the fractured vertebral body together again with great force, which results in a secondary burst fracture.
|
|
It is clear that a focus on the vertebral body status in the classification of thoracolumbar trauma poses diagnostic difficulties. Jeanneret et al (20), in describing a series of burst fractures combined with posterior ligamentous complex disruption, commented that "in contradistinction to the usual burst fractures, the upper fragment of the fractured posterior wall is not, or is only minimally retropulsed in relation to the vertebral body above." They named these "burst-shear flexion-distraction" injuries. Ferguson and Allen (4) found a high frequency of bone in the spinal canal, with 34 such cases in their series of 54 patients with thoracolumbar fractures. They reclassified thoracolumbar injuries on the basis of the mechanism of injury. In their "compressive flexion" subgroup, the posterosuperior margin of the vertebral body had rotated into the neural canal. This fracture pattern was associated with an increase in the posterior vertebral body wall height relative to adjacent vertebral bodies. These were therefore not classic burst fractures, in which the posterior vertebral body height decreases (9,11).
The study by Ferguson and Allen (4) was done without the benefit of MR imaging, and the authors were not able to comment on the integrity of the posterior ligamentous complex. Petersilge et al (21), using MR imaging, noted that 28% of their series of burst fractures had a torn supraspinous ligament, and they used the compressive flexion concept of Ferguson and Allen to account for these "unstable burst" injuries. The height of the middle column, however, was not evaluated as part of the radiographic assessment, nor was the integrity of the pedicles investigated in their patient series.
Terk et al (22) also found two subtypes of burst injuries, those with and those without posterior ligamentous injury, which they designated as unstable and stable, respectively. They made no observations about the pedicles or the posterior vertebral body height, but they did comment on the similarity of severe compression fracture in association with torn posterior ligamentous complex to the group of flexion-distraction injuries. It has been shown that radiographs are unsatisfactory for accurately discriminating between the subtypes of thoracolumbar injury classification (5–7). Oner et al (5,6) recommended that these injuries be reclassified.
It is hoped that an awareness of the spectrum of injury patterns that can arise in Chance-type trauma will increase the likelihood of an accurate diagnosis with MR imaging. MR imaging is essential not only for depicting the posterior osteoligamentous complex but also for verifying that a seemingly uncomplicated burst fracture is not part of a flexion-distraction injury. The label attached to the injury by the interpreting radiologist is important; "flexion-distraction" or "Chance" implies a high risk of associated visceral injury, which may be overlooked by the clinician if the term "unstable burst" is employed.
This study was retrospective and descriptive. The small number of patients limited the statistical analysis.
In conclusion, MR imaging is essential for the accurate understanding, documentation, and depiction of osseous and soft-tissue injuries that occur in Chance-type spinal injuries. It allows the evaluation that is necessary to conceptualize the mechanisms of injury fully. Respected authors and radiologists in practice have struggled to fit their observations into existing classification systems, but these systems need to evolve to encompass and reflect the findings of MR imaging.
|
FOOTNOTES
|
|---|
Abbreviations: SE = spin echo • STIR = short inversion time inversion-recovery
Authors stated no financial relationship to disclose.
Author contributions: Guarantor of integrity of entire study, V.N.C.P.; study concepts and design, C.J.G., V.N.C.P.; literature research, C.J.G., V.N.C.P.; clinical studies, all authors; data acquisition, all authors; data analysis/interpretation, C.J.G., V.N.C.P.; manuscript preparation, definition of intellectual content, and final version approval, C.J.G., V.N.C.P.; manuscript editing and revision/review, all authors
|
References
|
|---|
- Denis F. Spinal instability as defined by the three-column spine concept in acute spinal trauma. Clin Orthop 1984; 189:65–76.
- Anderson PA, Henley MB, Rivara FP, Maier RV. Flexion distraction and Chance injuries to the thoracolumbar spine. J Orthop Trauma 1991; 5:153–160.[Medline]
- Chance GQ. Note on a type of flexion fracture of the spine. Br J Radiol 1969; 42:452–453.[Abstract/Free Full Text]
- Ferguson RL, Allen BL Jr. A mechanistic classification of thoracolumbar spine fractures. Clin Orthop Relat Res 1984; 189:77–88.
- Oner FC, van Gils AP, Faber JA, Dhert WJ, Verbout AJ. Some complications of common treatment schemes of thoracolumbar spine fractures can be predicted with magnetic resonance imaging: prospective study of 53 patients with 71 fractures. Spine 2002; 27:629–636.[CrossRef][Medline]
- Oner FC, van Gils AP, Dhert WJ, Verbout AJ. MRI findings of thoracolumbar spine fractures: a categorisation based on MRI examinations of 100 fractures. Skeletal Radiol 1999; 28:433–443.[CrossRef][Medline]
- Lander P, Marchesi D, Al Obaid M. Reliability of the AO comprehensive classification of thoracolumbar fractures using routine radiographs. Presented at the European Congress of Radiology, Vienna, Austria, March 1–5, 2002.
- Denis R, Allard M, Atlas H, Farkouh E. Changing trends with abdominal injury in seatbelt wearers. J Trauma 1983; 23:1007–1008.[Medline]
- Campbell DK, Austin RF. Seat-belt injury: injury of the abdominal aorta. Radiology 1969; 92:123–124.[Medline]
- Woelfel GF, Moore EE, Cogbill TH, Van Way CW 3rd. Severe thoracic and abdominal injuries associated with lap-harness seatbelts. J Trauma 1984; 24:166–167.[Medline]
- Dehner JR. Seatbelt injuries of the spine and abdomen. Am J Roentgenol Radium Ther Nucl Med 1971; 111:833–843.[Medline]
- Rogers LF. The roentgenographic appearance of transverse or Chance fractures of the spine: the seat belt fracture. Am J Roentgenol Radium Ther Nucl Med 1971; 111:844–849.[Medline]
- Smith WS, Kaufer H. Patterns and mechanisms of lumbar injuries associated with lap seat belts. J Bone Joint Surg Am 1969; 51:239–254.[Abstract/Free Full Text]
- Lee HM, Kim HS, Kim DJ, Suk KS, Park JO, Kim NH. Reliability of magnetic resonance imaging in detecting posterior ligament complex injury in thoracolumbar spinal fractures. Spine 2000; 25:2079–2084.[CrossRef][Medline]
- Hoshikawa T, Tanaka Y, Kokubun S, Lu WW, Luk KD, Leong JC. Flexion-distraction injuries in the thoracolumbar spine: an in vitro study of the relation between flexion angle and the motion axis of fracture. J Spinal Disord Tech 2002; 15:139–143.[Medline]
- Frankel HL, Rozycki GS, Ochsner MG, Harviel JD, Champion HR. Indications for obtaining surveillance thoracic and lumbar spine radiographs. J Trauma 1994; 37:673–676.[Medline]
- Qaiyum M, Tyrrell PN, McCall IW, Cassar-Pullicino VN. MRI detection of unsuspected vertebral injury in acute spinal trauma: incidence and significance. Skeletal Radiol 2001; 30:299–304.[CrossRef][Medline]
- Gertzbein SD, Court-Brown CM. Flexion-distraction injuries of the lumbar spine. Mechanisms of injury and classification. Clin Orthop 1988; 227:52–60.
- Begeman P, King A, Prasad P. Spinal loads resulting from –Gx acceleration. In: Proceedings of the 17th Stapp Car Crash Conference. New York, NY: Society of Automotive Engineers, 1973; 343–360.
- Jeanneret B, Ho PK, Magerl F. Burst-shear flexion-distraction injuries of the lumbar spine. J Spinal Disord 1993; 6:473–481.[Medline]
- Petersilge CA, Pathria MN, Emery SE, Masaryk TJ. Thoracolumbar burst fractures: evaluation with MR imaging. Radiology 1995; 194:49–54.[Abstract/Free Full Text]
- Terk MR, Hume-Neal M, Fraipont M, Ahmadi J, Colletti PM. Injury of the posterior ligament complex in patients with acute spinal trauma: evaluation by MR imaging. AJR Am J Roentgenol 1997; 168:1481–1486.[Abstract]
This article has been cited by other articles:
|
|
|
|
 
M. P. Bernstein, S. E. Mirvis, and K. Shanmuganathan
Chance-Type Fractures of the Thoracolumbar Spine: Imaging Analysis in 53 Patients
Am. J. Roentgenol.,
October 1, 2006;
187(4):
859 - 868.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
