Thoracic and lumbar spinal column injury in adults: Evaluation

INTRODUCTION — Although less common than cervical spinal column injuries, injuries of the thoracolumbar (TL) spinal column are increasing globally [1]. While these injuries can be devastating themselves, they are also commonly associated with major internal injuries of the chest and abdomen following high-energy trauma.

The evaluation and classification of thoracic and lumbar spinal column injuries in adults is reviewed here.

●The evaluation and initial management of cervical spinal column injuries are discussed separately. (See "Cervical spinal column injuries in adults: Evaluation and initial management".)

●The epidemiology, spinal anatomy, common mechanisms of injury, classification schemes, and important types of TL spinal column injuries are discussed separately. (See "Spinal column injuries in adults: Types, classification, and mechanisms".)

●The anatomy, clinical presentation, classification, evaluation, and management of traumatic spinal cord injuries are discussed separately. (See "Acute traumatic spinal cord injury" and "Anatomy and localization of spinal cord disorders".)

●Pediatric spine injuries are discussed separately. (see "Evaluation and acute management of cervical spine injuries in children and adolescents" and "Overview of cervical spinal cord and cervical peripheral nerve injuries in the child or adolescent athlete")

EVALUATION FOR ASSOCIATED INJURIES — Patients with possible thoracolumbar (TL) spinal column injuries have often been involved in high-energy trauma (eg, motor vehicle collision, fall from a height) and sustained multiple injuries. Thus, the differential diagnosis for upper or lower back pain in these patients is broad. In addition to severe intrathoracic (eg, aortic injury), intra-abdominal, and pelvic injuries, clinicians should evaluate for spinal cord injury, spinal epidural hematoma, paraspinal hematoma, retroperitoneal hematoma and kidney injury, and soft tissue injuries. Diagnostic imaging must be obtained to differentiate among these entities. (See "Initial evaluation and management of blunt thoracic trauma in adults" and "Initial evaluation and management of blunt abdominal trauma in adults" and "Pelvic trauma: Initial evaluation and management" and "Acute traumatic spinal cord injury" and "Blunt genitourinary trauma: Initial evaluation and management".)

CLINICAL EVALUATION

General guidelines and risk factors — Once immediate, life-threatening conditions are addressed, a secondary survey of the trauma patient is performed to look closely for any sign of injury. In all cases of possible spinal injury, spinal immobilization must be maintained until the possibility of injury is excluded by clinical or radiographic means. General performance of the secondary survey is reviewed separately. (See "Initial management of trauma in adults", section on 'Secondary evaluation and management'.)

Thoracolumbar (TL) fractures most commonly occur after motor vehicle collisions, a fall from a height, or penetrating trauma, including gunshot wounds. Many forces may be involved in TL fractures, including compression, flexion, extension, rotation, shear, and distraction. However, the most common fracture mechanism is that of a compression fracture from an axial load. Age, female gender, and osteoporosis are considered major risk factors, as vertebral body compression fractures are more common in those with decreased bone density. Some malignancies have a propensity to metastasize to the spinal column (eg, lung, breast, prostate) and may lead to pathological fractures. (See "Spinal column injuries in adults: Types, classification, and mechanisms", section on 'Mechanisms of injury'.)

Spine immobilization — Although a long board is used to facilitate spinal immobilization of trauma patients in the prehospital setting, there is considerable variation in technique [2] and in the length of time a patient remains on the board after arriving at the emergency department (ED) [3]. Traditionally, spinal immobilization has included a long backboard, rigid cervical collar, and lateral head supports; however, some states and emergency medical services have revised their protocols to omit the backboard. Many experts advocate that the long board only be used for extrication and transport to the hospital, and that it be removed promptly after the primary survey [4]. We concur with this viewpoint.

Prolonged immobilization on a spine board can lead to decubitus ulcers, contribute to respiratory compromise, increase risk for aspiration, and may not adequately immobilize the spine (eg, the gap often visible between the patient's back and the board clearly demonstrates the lack of lumbar support) [3]. However, rigorous outcome studies to determine the effect of spinal immobilization upon the mortality, neurological morbidity, and overall outcomes of trauma patients are lacking [5].

The optimal technique for immobilizing the spine remains unclear. Several techniques and a range of immobilization devices are used, including standard backboards, vacuum splints, and scoop stretchers [6]. In most instances, we recommend the following approach: Once the primary survey is completed, a patient being managed with spine immobilization is log-rolled onto their side while axial alignment is maintained. At least three people are needed to perform the log-roll properly: one person maintains cervical spine immobilization, the second rolls the patient towards them, and the third inspects and palpates the entire length of the spine. If there is no sign of injury and no spinal tenderness, the backboard is removed. However, some clinicians prefer to continue immobilization with a long board until initial radiographs are obtained, provided these studies are not delayed. Note that some spine boards contain metal rods, which may interfere with radiographs and should be removed prior to imaging.

If continued immobilization is deemed necessary (eg, due to radiographic abnormality, neurologic deficit, or high suspicion for spinal injury in an obtunded patient), the patient is carefully log-rolled while maintaining axial alignment, the spine board is removed, and the patient is moved onto a firm mattress, generally using a sliding board [7]. Often, a slider board can be introduced at the same time the long board is removed, thereby minimizing patient motion. Nevertheless, the key concept is that even patients who warrant continued immobilization should be removed from the hard spine board before complications, such as pressure sores or aspiration, occur. When there is no concomitant vertebral fracture, the incidence of isolated ligamentous injury of the TL spine is low [8]. If an injury of the cervical spine is known or suspected, a cervical collar should be kept in place.

Clinical findings of thoracolumbar injury — As part of the secondary survey, the entire back, chest, and abdomen should be inspected and palpated. Patients with major injuries to the chest, abdomen, or pelvis often sustain injuries to the TL spine as well. Back pain and neurologic dysfunction, particularly involving the pelvis and lower extremities, are clinical findings associated with TL spinal column injuries. Other signs suggestive of spinal column injury include contusions, abrasions, lacerations, open wounds, and muscle spasm in the general area of the spine. Any deviations from normal spinal curvature should be noted. Fractures can cause a kyphotic or scoliotic deformity; muscle spasms can cause a straightening of the spine and loss of lordosis. Note that the region of the spine between the T11 and L4 vertebrae is most susceptible to injury. (See "Spinal column injuries in adults: Types, classification, and mechanisms", section on 'Anatomy'.)

Depending upon the body habitus of the injured patient, the ability to palpate any deformity or step-off of the spine may be difficult, particularly in an uncooperative patient. A prospective observational study reported that clinical examination detected only 27 of 56 TL spine fractures among blunt trauma patients capable of being evaluated (sensitivity 48.2 percent, specificity 84.9 percent) [9]. In this study, approximately two-thirds of the TL fractures sustained were of the transverse process. Thus, advanced imaging of the involved region is warranted if the clinician detects focal tenderness, a neurologic deficit, or spinal deformity. Furthermore, given the limited sensitivity and specificity of the physical examination for TL spine injury, the absence of tenderness or deformity should not dissuade the clinician from obtaining appropriate diagnostic imaging if they are concerned about possible injury based on mechanism or other known injuries the patient has sustained. (See 'Decision rules for imaging thoracolumbar injury' below.)

Whenever feasible, a focused but systematic neurologic evaluation should be performed in all patients with a possible TL spinal column injury. Ideally, this examination includes assessment of motor function, sensation, reflexes, and position sense. A significant portion of TL fractures damage the distal spinal cord or cauda equina to some degree [10-12], and such injuries can present with lower extremity paresis, lower extremity or saddle anesthesia, or loss of bladder or anal sphincter function.

Decreased rectal tone is thought to be a late finding of the cauda equina syndrome and is often absent immediately following an acute injury [13]. Although previous iterations of Advanced Trauma Life Support (ATLS) advocated performing a digital rectal examination on all trauma patients during the secondary survey, there is no clear evidence that the examination is an accurate means for detecting spinal cord injury [14-17]. The most recent edition of ATLS does not mandate a rectal examination during the secondary survey [18]. The neurologic examination and localization of spinal cord lesions are discussed separately. (See "The detailed neurologic examination in adults" and "Anatomy and localization of spinal cord disorders".)

Among patients who sustain high-energy trauma, scapular contusions suggest a rotation or flexion-rotation injury of the thoracic spine. Close evaluation of the thoracic spine, generally including spinal reconstructions from computed tomography (CT) of the chest, is prudent in such cases. Large contusions in the lumbosacral area following trauma are associated with shear fractures in this region. (See "Spinal column injuries in adults: Types, classification, and mechanisms", section on 'Translational spinal column injury'.)

The classic presentation for a flexion-distraction injury (eg, Chance fracture) of the TL spine (image 1 and image 2) is that of a backseat passenger restrained by a lap belt only, without a shoulder strap [19]. Such a patient may manifest ecchymosis on the lower abdomen, the so-called "seatbelt sign" (picture 1), and tenderness at the lumbar fracture site. However, the absence of these signs does not preclude the diagnosis. Although neurologic deficits occur in fewer than 5 percent, patients with Chance fractures often sustain intra-abdominal injury, such as intestinal perforation, hematoma, or contusion. (See "Spinal column injuries in adults: Types, classification, and mechanisms", section on 'Flexion-distraction (lap belt) injuries'.)

The pelvis and lower extremities should be examined, paying particular attention to pelvic stability, signs of injury, and neurologic function. Approximately 10 percent of calcaneal fractures are associated with lumbar fractures when the mechanism of injury involves axial compression, such as a fall from a significant height [20]. (See "Spinal column injuries in adults: Types, classification, and mechanisms", section on 'Burst fractures' and "Calcaneus fractures".)

Although performance of a careful physical examination remains important, the findings of several observational studies suggest that clinical examination alone is not sufficiently sensitive or specific to rule out or identify significant TL spine injuries [9,21,22]. Several studies support this conclusion:

●According to a retrospective observational study of 536 blunt trauma patients with TL spine fractures, the absence of back pain or tenderness during a log-roll examination did not exclude a clinically significant thoracic fracture even among patients with a Glasgow Coma Scale (GCS) score of 15 who were not under the influence of alcohol and had not yet been treated with opioids [22].

●In a large prospective study, over 20 percent of patients with a TL spine injury requiring surgical management or fixed immobilization had no significant findings on physical examination [23]. In this study, clinical examination findings (eg, midline tenderness, deformity, neurologic deficit) demonstrated a sensitivity and specificity for TL spine injury of 78.4 and 72.9 percent, respectively.

Thus, clinical examination alone is insufficient to determine the need for TL spine imaging. Other factors, such as mechanism of injury and age, in addition to examination findings, must be considered. (See 'Decision rules for imaging thoracolumbar injury' below.)

DECISION RULES FOR IMAGING THORACOLUMBAR INJURY

When is imaging necessary? — In contrast to cervical spine trauma, few studies have been performed to develop decision rules for imaging potential thoracic and lumbar spine injuries. Based on the available evidence, we suggest that imaging studies of the thoracolumbar (TL) spine be obtained for trauma patients who meet any of the following criteria [23-30]:

●Signs of TL spine injury, including:

•Focal pain or tenderness over the TL spine

•Signs of injury (eg, bruising, hematoma, palpable step-off) along the TL spine

•Neurologic deficit consistent with TL injury

●High-force mechanism, including:

•Fall greater than 3 m (10 feet)

•Ejection from a vehicle

•Moderate- or high-velocity motor vehicle collision

•Unenclosed motor vehicle collision (eg, motorcycle collision)

•Automobile versus pedestrian

•Forceful direct blow (eg, struck with bat in area of TL spine)

●Presence of another spine injury, particularly a known cervical fracture [28,31]

●Painful distracting injury in patients with complaints or a mechanism consistent with TL spine injury

●Age over 60 years in patients with complaints or a mechanism consistent with TL spine injury (mechanism may include fall from standing in frail older adults) [23]

●Depressed mental status, including clinical signs of intoxication or Glasgow Coma Scale (GCS) score below 15, in patients with complaints or a mechanism consistent with TL spine injury

We concur with the general approach described in the Eastern Association for the Surgery of Trauma practice guidelines and believe that diagnostic imaging of the TL spine is generally not necessary in blunt trauma patients with a normal mental status and none of the risk factors described above [28].

The presence of one vertebral fracture in a blunt trauma patient mandates radiographic evaluation of the entire spine, as the risk of a second, noncontiguous fracture may be as high as 20 percent [32,33]. The association with a second fracture appears to be higher in patients who sustain high-energy blunt trauma and those with cervical fractures. Using data from the National Trauma Database of the United States, a retrospective analysis that included over 80,000 patients with spine fractures following blunt trauma reported that 13 percent of patients with a cervical spine fracture also had either a thoracic or a lumbar fracture, while 10 percent of patients had concomitant thoracic and lumbar fractures.

It is important to note the limitations of the physical examination in the setting of TL spine injury. In one large prospective study, over 20 percent of patients with a TL spine injury requiring surgical management or fixed immobilization had no significant findings on physical examination [23]. Diagnostic imaging may still be required despite an unremarkable physical examination. (See 'Clinical findings of thoracolumbar injury' above.)

Trauma patients with altered mental status may need imaging depending upon their level of alertness, examination findings, and clinical suspicion. Such patients are at higher risk of injury. Spinal immobilization is maintained until unstable injury is ruled out by clinical or radiologic evaluation. The manual for Advanced Trauma Life Support (ATLS) describes a similar approach, recommending imaging for "any patient with an altered level of consciousness or cognitive dysfunction, GCS <15, multisystem injuries, or a palpable gap or tenderness in the TL area" [34]. In addition, patients at greater risk for fracture due to osteoporosis or cancer with bone metastases may warrant imaging.

Computed tomography preferred over plain radiographs — For patients with major trauma, high-risk features, or depressed mental status, multidetector computed tomography (CT) is the imaging modality of choice to assess the TL spine (image 3). The reported sensitivity of CT for detecting thoracic and lumbar spine fractures is 94 to 100 percent; standard anteroposterior (AP) and lateral radiographs of the thoracic and lumbar spine have a sensitivity of 49 to 62 percent for thoracic and 67 to 82 percent for lumbar spine fractures [35,36].

In most instances, reconstructed images obtained from CT scans of the chest and abdomen performed as part of the patient's trauma evaluation can be used to assess the TL spine. The use of reconstructed images using a collimated field of view and "bone" kernel reduces time, cost, and radiation exposure. A retrospective comparison of 72 patients who underwent both abdomino-pelvic CT and an additional TL spine CT demonstrated no difference in fracture detection rates [37].

AP and lateral plain radiographs of the thoracic and lumbar spine are generally sufficient to assess for the presence of gross injury in patients at lower risk; however, the sensitivity of standard plain chest radiographs for TL spine injury is poor [38]. If plain spine radiographs reveal abnormalities, CT imaging is then obtained of areas with abnormal findings. Unfortunately, a review of 69 publications describing blunt injuries to the TL spine failed to clearly define the criteria to determine who should get plain radiographs versus CT scans [39]. It is best to obtain CT imaging if concern for a spine injury persists based on clinical impression or findings on plain radiographs.

Evidence and proposed decision rules — The best evidence available to guide decision-making about diagnostic imaging in patients with suspected TL injury is from a prospective study performed at 13 trauma centers in the United States that included 499 patients with TL injuries, 264 of whom required surgical management (29.2 percent) or rigid spine immobilization (TL spine orthosis; 70.8 percent) [23]. Injuries were identified by multidetector CT in most instances (93.3 percent). The study excluded patients with a GCS score below 15 and those who could not be assessed reliably due to intoxication or significant distracting injury. According to the study, the presence of all three of the following major risk factors predicted a clinically significant injury with a sensitivity of 98.9 percent and a specificity of 29 percent, and predicted the need for surgery with a sensitivity of 100 percent and a specificity of 27.3 percent:

●Suggestive clinical examination finding, including:

•TL spine pain

•Midline TL spine tenderness

•TL spine bony deformity

•Neurologic deficit

●Age ≥60 years

●High-risk mechanism, including:

•Fall from height

•Crush injury

•Motor vehicle collision with ejection or rollover

•Unenclosed vehicle (eg, motorcycle) collision

•Automobile versus pedestrian

Prospective validation studies of this decision rule are needed.

Other proposed decision rules demonstrate high sensitivity but limited specificity [29,30]. One group of researchers prospectively studied a cohort of 2404 patients undergoing TL spine radiographs following blunt trauma in order to assess six potential predictors of injury [29]:

●Complaints of TL spine pain

●TL spine tenderness

●Decreased level of consciousness

●Intoxication with alcohol or drugs

●Neurologic deficits

●Presence of a painful, distracting injury

The presence of any of the six predictors identified all 152 patients with TL fractures, demonstrating sensitivity of 100 percent (95% CI 98-100), but specificity was poor (3.9 percent; 95% CI 3.1-4.8).

A single-center, prospective study involving 58 thoracic spine injuries reported that a decision instrument using three variables (thoracic spine tenderness, altered mental status, and distracting painful injury) demonstrated high sensitivity (96.6 percent; 95% CI 88.1-99.6) and a high negative likelihood ratio (0.07, 95% CI 0.02-0.28) for a thoracic spine injury [40]. Lumbar spine injuries were not studied.

Another research group studied a potential decision rule for TL imaging by performing a chart review in which 100 subjects with TL spine fractures were compared with 100 randomly selected multitrauma patients and reported findings and test characteristics consistent with other groups [30].

COMPUTED TOMOGRAPHY FOR THORACOLUMBAR INJURY — Computed tomography (CT) is more accurate than plain radiograph for diagnosing thoracolumbar (TL) spinal column injury [28,41-43]. For patients who have sustained significant blunt TL trauma based upon mechanism or clinical findings and for obtunded patients, we suggest CT imaging be used to assess the TL spine (image 3). Reformatted images from standard CT studies of the chest and abdomen using bone kernel and collimated field of view provide accurate images without the need for additional radiation exposure and require minimal time to produce. Some trauma radiologists think that reformatted images are better than conventional images for diagnosing TL spine injuries, and we prefer this approach when resources permit [44].

Several studies support this approach:

●A systematic review found superior sensitivity and interobserver reliability in detecting TL spine injury with reformatted CT images compared with plain radiographs [45]. Furthermore, CT did not require further patient transportation, radiation exposure, cost, or time when reformatted images were used.

●A prospective evaluation of 222 consecutive trauma patients requiring TL spine imaging because of clinical findings or altered mental status found the accuracy of helical CT for TL fractures to be 99 percent (95% CI 96-100) while the accuracy of plain radiographs was 87 percent (95% CI 82-92) [46].

Images obtained from standard CT studies may be sufficient to assess for TL spine fractures. This was shown in an observational study in which 78 of 80 fractures of the TL spine were correctly identified using images from chest, abdomen, and pelvis CTs without reformatting [47]. Both missed injuries were minor fractures involving the transverse process in patients with multiple other spinal injuries.

Analyses of cost effectiveness comparing CT with plain radiographs of the TL spine report that although CT has a higher initial fixed cost, this is offset by a reduced need for repeat radiographic examinations due to inadequate studies, lower medicolegal costs (including costs of prolonged hospitalizations, rehabilitation, lost productivity, and malpractice suits from missed injuries), and fixed personnel costs [48,49].

A CT obtained for the evaluation of chest and abdominal injury provides sufficient data to screen for spinal fractures, thereby decreasing the time and cost of spine injury evaluation:

●In a retrospective review of 3537 blunt trauma patients, of whom 236 (7 percent) sustained a cervical, thoracic, or lumbar fracture, researchers found that CT identified 99.3 percent of all fractures. The one cervical and one thoracic fracture missed by CT required no or minimal (rigid cervical collar) treatment [50].

●A registry-based retrospective analysis of 573 trauma patients identified 54 who sustained spinal column injury and found CT to be 100 percent sensitive for detecting these injuries compared with 70 percent sensitivity for plain radiograph [51].

Should stand-alone imaging of the TL spine be necessary, it is important to be aware of relative radiation exposures. CT of the TL spine involves approximately 6 mSv of radiation exposure (range of 1.5 to 10 mSv, depending upon technique and other factors) compared with approximately 1.5 mSv (range of 0.5 to 1.8 mSv) using plain radiographs [52]. However, radiation exposure may be higher if multiple sets of plain radiographs are obtained or CT is performed following plain radiographs [52,53].

MAGNETIC RESONANCE IMAGING FOR THORACOLUMBAR INJURY — Magnetic resonance imaging (MRI) can be used to evaluate the integrity of the spinal ligaments, surrounding soft tissues, and for spinal hematomas in a patient with spinal column trauma [54]. However, vertebral fractures are better characterized by computed tomography (CT), and MRI examination is often felt to be unnecessary for the management and treatment of specific fractures [55]. When CT or plain radiographs demonstrate a thoracolumbar (TL) burst fracture, MRI is often used to characterize the integrity of the posterior ligamentous complex [56]. MRI is necessary to assess the spinal cord of any trauma patient with neurologic deficits or symptoms that suggest spinal cord injury [28]. (See "Acute traumatic spinal cord injury", section on 'Imaging'.)

THORACOLUMBAR INJURY CLASSIFICATION SYSTEMS/SCORES — There remains no universally accepted classification system or score for thoracolumbar (TL) spine injuries that would facilitate communication and help to standardize approaches to treatment. Some of the more clinically relevant scores include the Thoracolumbar Injury Classification and Severity Score (TLICS) and the Thoracolumbar AOSpine Injury Score (TL AOSIS). These scores are useful guides, but neither score perfectly predicts which patients ultimately require surgery; the TLICS may work better in children and the TL AOSIS in adults [57,58].

The distinction between major and minor fracture patterns using the three-column scheme (figure 1) and the five basic injury patterns are discussed separately. (See "Spinal column injuries in adults: Types, classification, and mechanisms", section on 'Thoracolumbar injury schemes and fracture patterns'.)

●Thoracolumbar Injury Classification and Severity Score (TLICS) – This was proposed to delineate neurologic and spinal stability [59,60]. The score is calculated based upon injury morphology, integrity of the posterior ligamentous complex, and neurologic status and has demonstrated good reliability and reproducibility [61]. Scoring of the TLICS is as follows:

•Injury morphology

-Compression = 1 point

-Burst = 1 point

-Translational/rotational = 3 points

-Distraction = 4 points

•Neurological status

-Intact = 0 points

-Nerve root injury = 2 points

-Spinal cord incomplete injury = 3 points

-Spinal cord complete injury = 2 points

-Cauda equina injury = 3 points

•Posterior ligament complex

-Intact = 0 points

-Injury suspected/indeterminate = 2 points

-Injured = 3 points

The total numerical score is used to guide treatment. A TLICS ≥5 suggests instability and the need for operative treatment, whereas a score ≤3 suggests stability. A score of 4 is considered indeterminate and either operative or conservative management may be indicated [62]. The TLICS fails to provide a treatment strategy for the neurologically intact patient with a TL burst fracture and possible posterior ligamentous complex disruption (thus having a score of 4).

Alternative classification schemes have been proposed, including the modified TLICS (mTLICS) [63]. In a small retrospective study, the mTLICS had better test characteristics than the original version [63].

●Thoracolumbar AOSpine Injury Score (TL AOSIS) – The AOSpine Thoracolumbar Spine Injury Classification system is based upon injury groups (compression, tension band, or translation), neurologic status, and clinical modifiers (eg, patient-specific comorbidities, such as ankylosing spondylitis or polytrauma) [64]. In several retrospective studies, the AOSpine system showed moderate inter-rater reliability and better reliability compared with the TLICS for identifying fracture morphology when applied by spine surgeons [65,66]. Prospective validation of this classification is needed [67].

The TL AOSIS was developed to recommend treatment strategies based upon the AOSpine classification system [58,64,68]. A TL AOSIS >5 suggests the need for operative treatment, a score ≤3 suggests conservative treatment, and a score of 4 or 5 is considered indeterminate. Scoring of the TL AOSIS is as follows:

•Compression fractures

-No injury/process fracture = 0 points

-Wedge/impaction = 1 point

-Split/pincer type = 2 points

-Incomplete burst = 3 points

-Complete burst = 5 points

•Tension band injury

-Posterior transosseous disruption = 5 points

-Posterior ligamentous disruption = 6 points

-Anterior ligamentous disruption = 7 points

-Translation injuries = 8 points

•Neurologic status

-Neurologically intact = 0 points

-Transient neurologic deficit = 1 point

-Symptoms or signs of radiculopathy = 2 points

-Incomplete spinal cord injury or cauda equina injury = 4 points

-Complete spinal cord injury = 4 points

-Neurologic examination is unobtainable = 3 points

•Case-specific modifiers

-Fractures with an indeterminate injury to the tension band based on spinal imaging, such as magnetic resonance imaging (MRI) or clinical examination = 1 point

-Patient-specific comorbidity, which might argue either for or against surgery for those patients with relative indications for surgery = 0 points

Studies comparing the TLICS and TL AOSIS have found good correlation with predicting which patients will require surgery, but the evidence is limited and retrospective. In a single-center study of 110 adults with a TL injury, the TL AOSIS more accurately predicted which patients had surgery compared with the TLICS (98 versus 87 percent) [58]. However, in a study of 54 children with traumatic TL fractures that were managed surgically, the TLICS performed much better compared with the TL AOSIS (TLICS ≥5: 87 percent versus TL AOSIS >5: 46 percent) [57].

DISPOSITION — The disposition of patients with spinal column injury depends primarily upon fracture stability and concomitant injuries. The goals of operative and nonoperative orthopedic intervention in spinal column injuries are to decompress neural structures, stabilize unstable segments, and restore alignment of the vertebrae [69]. Nonoperative management of thoracic and lumbar spinal fractures involves the use of a spinal orthotic vest or brace.

If a spinal column injury is deemed unstable, hospital admission and spine surgery consultation is mandatory. If consultation is not available onsite, immediate transfer must be arranged to a center that can provide these services. Patients with unstable fractures have commonly sustained multisystem trauma, and the extent of their other injuries determines whether they require admission to an intensive care unit or other monitored setting. (See "Initial management of trauma in adults" and "Overview of inpatient management of the adult trauma patient".)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: General issues of trauma management in adults" and "Society guideline links: Thoracic and lumbar spine injury in adults".)

SUMMARY AND RECOMMENDATIONS

●Background and causes – Injuries of the thoracolumbar (TL) spinal column are less common than cervical injuries but can produce devastating neurologic damage. They are commonly associated with major injuries of the chest, abdomen, and/or pelvis following high-energy trauma. Motor vehicle collision is the most common cause. (See "Spinal column injuries in adults: Types, classification, and mechanisms" and 'Evaluation for associated injuries' above.)

●Spinal immobilization – In all cases of possible spinal column injury, spinal immobilization must be maintained until the possibility of injury is excluded by clinical or radiographic means. However, such patients may be removed from the hard long spine board onto a firm, flat mattress using the log-roll technique, as prolonged immobilization on the long board is associated with aspiration and pressure sores. (See 'Spine immobilization' above.)

●Clinical findings – Back pain, back deformity, and neurologic dysfunction are among the most common clinical findings associated with TL spinal column injuries. A significant portion of TL fractures damage the distal spinal cord or cauda equina to some degree, and such injuries can present with lower extremity paresis, lower extremity or saddle anesthesia, or loss of bladder or rectal continence. Other signs suggestive of injury include contusions, abrasions, lacerations, open wounds, or muscle spasm in the general area of the spine. Of note, a significant number of patients with fractures requiring intervention may not manifest clinical findings. (See 'Clinical evaluation' above.)

●Indications for diagnostic imaging – Effective decision rules to determine the need for imaging in patients with possible TL spinal column injury await development. Until better evidence is available to guide practice, we believe it is reasonable to obtain radiographic studies of the thoracic and/or lumbar spine when any one of the following high-risk criteria exists (see 'Decision rules for imaging thoracolumbar injury' above):

•Focal pain or tenderness over the TL spine

•Signs of injury (eg, step-off, bruising, hematoma) along the TL spine

•Neurologic deficit consistent with TL injury

•High-force mechanism (eg, fall greater than 3 m [10 feet], ejection from a vehicle, moderate- or high-velocity motor vehicle collision, auto versus pedestrian)

•Painful distracting injury (particularly when elements of the mechanism or clinical evaluation suggest increased risk for TL injury)

•Presence of another spine injury, particularly a cervical spine fracture

●Selection of imaging study – For patients with major trauma, high-risk features, or depressed mental status, computed tomography (CT) is the preferred study to assess the TL spine. When available, reconstructed images obtained from multidetector CT scans of the chest, abdomen, and pelvis performed as part of the patient's general trauma evaluation should be used. Anteroposterior (AP) and lateral plain radiographs of the thoracic and lumbar spine are generally sufficient to assess for signs of gross injury in patients at low risk. CT imaging is then obtained of areas with abnormal findings. In addition, CT may be performed if clinical concern for injury persists despite unremarkable plain radiographs. Standard plain chest radiographs show poor sensitivity for TL spine injury. (See 'Computed tomography for thoracolumbar injury' above.)