Acute traumatic spinal cord injury

INTRODUCTION — Spinal cord injury has become epidemic in modern society. Despite advances made in the understanding of the pathogenesis and improvements in early recognition and treatment, it remains a devastating event, often producing severe and permanent disability. With a peak incidence in young adults, traumatic spinal cord injury (TSCI) remains a costly problem for society; direct medical expenses accrued over the lifetime of one patient range from 500,000 to 2 million US dollars [1].

This topic reviews acute TSCI. The anatomy and clinical localization of spinal cord disease, other diseases affecting the spinal cord, and the chronic complications of spinal cord injury are discussed separately. (See "Anatomy and localization of spinal cord disorders" and "Disorders affecting the spinal cord" and "Chronic complications of spinal cord injury and disease".)

Issues regarding injury to the vertebral column and ligaments are also discussed separately. (See "Spinal column injuries in adults: Types, classification, and mechanisms" and "Cervical spinal column injuries in adults: Evaluation and initial management" and "Thoracic and lumbar spinal column injury in adults: Evaluation".)

EPIDEMIOLOGY — Most demographic and epidemiologic data related to TSCI in the United States have been collected by the Model Spinal Cord Injury Care Systems and are published by the National Spinal Cord Injury Statistical Center [2]. In the United States, the incidence of TSCI in 2010 was approximately 40 per million persons per year, or approximately 12,400 annually [3], with approximately 250,000 living survivors of TSCI in the United States in July 2005. Similar figures are reported in Canada [4]. The incidence in the United States is higher than in most other countries.

The causes of TSCI in the United States are [3]:

●Motor vehicle accidents: 48 percent

●Falls: 16 percent

●Violence (especially gunshot wounds): 12 percent

●Sports accidents: 10 percent

●Other: 14 percent

Statistics differ somewhat in other countries. In Canada and western Europe, TSCI due to violence is rare, while in developing countries, violence is even more common [5,6]. Soldiers deployed in armed conflicts also have a substantial risk of TSCI [7].

Risk factors for TSCI have been identified. Prior to 2000, the most frequent victim was a young male with a median age of 22. Since that time, the average age has increased in the United States to 37 years in 2010 [3], presumably as a reflection of the aging population. Males continue to make up 77 to 80 percent of cases [2,3,5,6,8]. Alcohol plays a role in at least 25 percent of TSCI [1,9]. Underlying spinal disease can make some patients more susceptible to TSCI [1,10]. These conditions include:

●Cervical spondylosis

●Atlantoaxial instability

●Congenital conditions, eg, tethered cord

●Osteoporosis

●Spinal arthropathies, including ankylosing spondylitis or rheumatoid arthritis (see "Clinical manifestations of axial spondyloarthritis (ankylosing spondylitis and nonradiographic axial spondyloarthritis) in adults", section on 'Neurologic manifestations' and "Cervical subluxation in rheumatoid arthritis")

PATHOPHYSIOLOGY — Most spinal cord injuries are produced in association with injury to the vertebral column. These can include any one or more of the following [1]:

●Fracture of one or more of the bony elements

●Dislocation at one or more joints

●Tearing of ligament(s)

●Disruption and/or herniation of the intervertebral disc

The injury reflects the force and direction of the traumatic event and subsequent fall, which produces pathologic flexion, rotation, extension, and/or compression of the spine, as well as the anatomic vulnerability of individual spinal elements. Most vertebral injuries in adults involve both fracture and dislocation [1]. The type of injury has implications for the stability of the spinal column and the risk for further spinal cord injury (table 1). (See "Spinal column injuries in adults: Types, classification, and mechanisms".)

The mechanisms surrounding injury to the spinal cord itself are often discussed in terms of primary and secondary injury. The primary injury refers to the immediate effect of trauma, which includes forces of compression, contusion, and shear injury to the spinal cord. In the absence of cord transection or frank hemorrhage (both relatively rare in nonpenetrating injuries), the spinal cord may appear pathologically normal immediately after trauma. Penetrating injuries (eg, knife and gunshot injuries) usually produce a complete or partial transection of the spinal cord. An increasingly described phenomenon, however, is a spinal cord injury following a gunshot wound that does not enter the spinal canal [11]. Presumably, the spinal cord injury in these cases results from kinetic energy emitted by the bullet.

A secondary, progressive mechanism of cord injury usually follows, beginning within minutes and evolving over several hours after injury [1,12-15]. The processes propagating this phenomenon are complex and incompletely understood [16,17]. Possible mechanisms include ischemia, hypoxia, inflammation, edema, excitotoxicity, disturbances of ion homeostasis, and apoptosis [1,16,18]. The phenomenon of secondary injury is sometimes clinically manifest by neurologic deterioration over the first 8 to 12 hours in patients who initially present with an incomplete cord syndrome.

As a result of these secondary processes, spinal cord edema develops within hours of injury, becomes maximal between the third and sixth day after injury, and begins to recede after the ninth day. This is gradually replaced by a central hemorrhagic necrosis [19].

CLINICAL PRESENTATION — A patient with a cord injury typically has pain at the site of the spinal fracture. This is not always a reliable feature to exclude TSCI. Patients with TSCI often have associated brain and systemic injuries (eg, hemothorax, extremity fractures, intra-abdominal injury) that may limit the patient's ability to report localized pain [1,8]. These also complicate the initial evaluation and management of patients with TSCI and affect prognosis.

Approximately half of TSCIs involve the cervical cord and as a result present with quadriparesis or quadriplegia [2,6]. The severities of cord syndromes are classified using the American Spinal Injury Association (ASIA) Scale (table 2) [20].

Complete cord injury — In a complete cord injury (ASIA grade A), there will be a rostral zone of spared sensory levels (eg, the C5 and higher dermatomes spared in a C5-6 fracture-dislocation), reduced sensation in the next caudal level, and no sensation in levels below, including none in the sacral segments, S4-S5. Similarly, there will be reduced muscle power in the level immediately below the injury, followed by complete paralysis in more caudal myotomes. In the acute stage, reflexes are absent, there is no response to plantar stimulation, and muscle tone is flaccid. A male with a complete TSCI may have priapism. The bulbocavernosus reflex is usually absent. Urinary retention and bladder distension occur. (See "Anatomy and localization of spinal cord disorders", section on 'Segmental syndrome'.)

Incomplete injury — In incomplete injuries (ASIA grades B through D), there are various degrees of motor function in muscles controlled by levels of the spinal cord caudal to the injury. Sensation is also partially preserved in dermatomes below the area of injury. Usually, sensation is preserved to a greater extent than motor function because the sensory tracts are located in more peripheral, less vulnerable areas of the cord. The bulbocavernosus reflex and anal sensation are often present.

The relative incidence of incomplete versus complete spinal cord injury has increased over the last half century [1]. This trend has been attributed to improved initial care and retrieval systems that emphasize the importance of immobilization after injury. (See 'Initial evaluation and treatment' below.)

Central cord syndrome — An acute central cord syndrome, characterized by disproportionately greater motor impairment in upper compared with lower extremities, bladder dysfunction, and a variable degree of sensory loss below the level of injury, is described after relatively mild trauma in the setting of preexisting cervical spondylosis [21,22]. (See "Cervical spondylotic myelopathy", section on 'Clinical presentation' and "Cervical spondylotic myelopathy", section on 'Patients with acute deterioration' and "Anatomy and localization of spinal cord disorders", section on 'Central cord syndromes'.)

Anterior cord syndrome — Lesions affecting the anterior or ventral two-thirds of the spinal cord, sparing the dorsal columns, usually reflect injury to the anterior spinal artery. When this occurs in TSCI, it is believed that this more often represents a direct injury to the anterior spinal cord by retropulsed disc or bone fragments rather than primary disruption of the anterior spinal artery. (See "Anatomy and localization of spinal cord disorders", section on 'Ventral (anterior) cord syndrome'.)

Transient paralysis and spinal shock — Immediately after a spinal cord injury, there may be a physiologic loss of all spinal cord function caudal to the level of the injury, with flaccid paralysis, anesthesia, absent bowel and bladder control, and loss of reflex activity [23,24]. In males, especially those with a cervical cord injury, priapism may develop. There may also be bradycardia and hypotension not due to causes other than the spinal cord injury. This altered physiologic state may last several hours to several weeks and is sometimes referred to as spinal shock.

We believe that this loss of function may be caused by the loss of potassium within the injured cells in the cord and its accumulation within the extracellular space, causing reduced axonal transmission. As the potassium levels normalize within the intracellular and extracellular spaces, this spinal shock wears off. Clinical manifestations may normalize but are more usually replaced by a spastic paresis reflecting more severe morphologic injury to the spinal cord.

A transient paralysis with complete recovery is most often described in younger patients with athletic injuries. These patients should undergo evaluation for underlying spinal disease before returning to play.

INITIAL EVALUATION AND TREATMENT

In the field — The primary assessment of a patient with trauma in the field follows the ABCD prioritization scheme: Airway, Breathing, Circulation, Disability (neurologic status). If the patient has a head injury, is unconscious or confused, or complains of spinal pain, weakness, and/or loss of sensation, then a traumatic spinal injury should be assumed. Extreme care should be taken to allow as little movement of the spine as possible to prevent more cord injury. Techniques to minimize spine movement include the use of log-roll movements and a backboard for transfer and placement of a rigid cervical collar [25].

In the emergency department — Management in the emergency department continues to prioritize assessment and stabilization following the ABCD scheme. Life-threatening priorities related to other injuries, such as systemic bleeding, breathing difficulties, or a pneumothorax, can take precedence over the spinal cord injury. (See "Initial management of trauma in adults".)

●Vital signs including heart rate, blood pressure, respiratory status, and temperature require ongoing monitoring. Capnography can provide a useful method of monitoring respiratory status in the emergency department. (See "Carbon dioxide monitoring (capnography)".)

●The patient with a high cervical cord injury may breathe poorly and may require airway suction or intubation. Respiratory mechanical support may be needed; approximately one-third of patients with cervical injuries require intubation within the first 24 hours [26]. Rapid-sequence intubation with in-line spinal immobilization is the preferred method when an airway is urgently required. If time is not an issue, intubation over a flexible fiberoptic laryngoscope may be a safer, effective option. (See "Rapid sequence intubation in adults for emergency medicine and critical care".)

●Hypoxia in the face of cord injury can adversely affect neurologic outcome. Arterial oxygenation should be monitored and supplemented as needed.

●Hypotension may occur due to blood loss from other injuries or due to blood pooling in the extremities lacking sympathetic tone because of the disruption of the autonomic nervous system (neurogenic shock). Prolonged hypoperfusion may adversely affect prognosis. Elevation of the legs, the head-dependent position, blood replacement, and/or vasoactive agents may be required.

●Until spinal injury has been ruled out (see 'Imaging' below), immobilization of the neck and body must be maintained using cervical collar, straps, tape, and blocks. Athletic headgear should be left on.

●A neurologic examination should be completed as soon as possible to determine the level and severity of the injury, both of which impact prognosis and treatment (see "Anatomy and localization of spinal cord disorders", section on 'Clinical localization'). An evaluation of mental status and cranial nerve function should be included, as many patients with TSCI have also suffered a head injury.

●The patient must be checked for bladder distension by palpation or ultrasound. A urinary catheter should be inserted as soon as possible, if not done previously, to avoid harm due to bladder distension.

IMAGING — Cervical spine imaging is often performed in trauma patients regardless of suspected TSCI. Patients who present with symptoms of TSCI require imaging, typically with computed tomography (CT), to show bone damage. If magnetic resonance imaging (MRI) is available, provided the spine is stabilized, MRI can be performed to show the extent of spinal cord damage, since the spinal cord is rarely well seen using CT scanning.

Screening assessments — While a full set of cervical spine films was traditionally required on all trauma patients before a cervical collar could be removed, patients are now stratified into high- and low-risk categories based on clinical decision rules. Patients who are not clinically evaluable for TSCI because of obtundation or confusion are assumed to have a TSCI until proven otherwise. Indications for screening imaging studies and the appropriate choice of testing are discussed in detail separately. (See "Suspected cervical spine injury in adults: Choice of imaging" and "Cervical spinal column injuries in adults: Evaluation and initial management".)

Plain radiographs — After neurologic examination, plain radiographs provide a rapid assessment of alignment, fractures, and soft tissue swelling and are, in general, the first method of assessment of suspected traumatic vertebral and/or spinal cord injury. A complete set of cervical radiographs includes anteroposterior, lateral, and open-mouth odontoid views. Oblique views may be necessary if one suspects a lateral mass or facet injury or damage. All cervical vertebrae and the top of the T1 vertebra should be visualized if possible. In muscular males with a neck injury, pulling the shoulders down by pulling down on the wrists in a straight line and downward towards the feet may better allow visualization of the lower cervical vertebrae. A swimmer's view should be performed if the lower cervical levels and the top of T1 are not adequately visualized. While there are reports of missed cervical spine injury with plain films, it is rare to miss significant injuries with adequate performance and interpretation of plain films of the occiput through the top of T1 [27,28].

Neurologic signs and symptoms of cervical spine injury in the setting of normal plain radiographs warrant further imaging studies.

Patients who have pain in the thoracic or lumbar areas, especially with an appropriate neurologic deficit, also require lateral, anteroposterior, and sometimes oblique plain radiographs of either the thoracic spine, lumbar region, or both. Such spinal injuries, especially with a neurologic deficit, require further imaging.

Computed tomography — Prospective case series report a higher sensitivity of helical CT for detecting spinal fracture when compared with plain radiographs; this is particularly true for cervical spine fracture [28-33]. This study can also be done without moving the patient out of the supine position. When a head CT is required to rule out head injury, it may be most cost and time efficient to use CT of the head as part of the initial imaging study of the neck as well.

All abnormalities on screening plain films or CT are followed up with a more detailed CT scan of the area in question, with fine, 2 mm cuts as needed. Areas not well visualized on plain films should be further imaged as well. This test is very sensitive for defining bone fractures in the spine. Because CT is more sensitive than plain films, patients who are suspected to have a spinal injury and have normal plain films should also undergo CT. CT also has advantages over plain films in assessing the patency of the spinal canal. CT also provides some assessment of the paravertebral soft tissues and perhaps of the spinal cord as well, but is inferior in that regard to MRI.

Myelography — When MRI is available, myelography with soluble contrast media is rarely if ever used, but remains an alternative in combination with CT when MRI cannot be performed and spinal canal compromise is suspected.

Magnetic resonance imaging — The indications for MRI in the evaluation of acute TSCI have not been defined [34,35].

The chief advantage of MRI is that it provides a detailed image of the spinal cord as well as spinal ligaments, intervertebral discs, and paraspinal soft tissues that is superior to CT and is more sensitive for detecting epidural hematoma [34,36-39]. CT, however, is better than MRI in assessing bony structures. In the absence of a cord transection or intramedullary hemorrhage, MRI is not perfectly sensitive to cord damage in the earliest stages of TSCI. MRI has other disadvantages: it is contraindicated in the setting of a cardiac pacemaker and metallic foreign bodies, life support equipment may be incompatible with the performance of MRI, and the patient is enclosed during the study, which can pose some risk for monitoring vital signs and for maintaining an airway. In some centers, MRI is not always available because of resource and personnel issues.

Nonetheless, if the patient's clinical status permits, MRI can provide valuable information that complements CT regarding the extent and mechanism of spinal cord injury, which can influence treatment and prognosis [34,40,41]. MRI is also indicated in patients with negative CT scan who are suspected to have TSCI, in order to detect occult ligamentous or disc injury or epidural hematoma [42]. In a systematic review of reported case series, 5.8 percent of individuals with negative CT scan who went on to have an MRI were found to have a traumatic spine injury [43]. While it has been suggested that nonalert patients require MRI in addition to CT to exclude TSCI, one case series suggests that if obtunded patients are observed to have grossly normal motor movement in all extremities, CT scan is sufficient in this population [44].

Spinal cord injury without radiographic abnormality — A category of TSCI called spinal cord injury without radiographic abnormality (SCIWORA) originated prior to the use of MRI and referred to patients with a myelopathy without evidence of traumatic vertebral injury on plain radiographs or CT. Because MRI provides superior imaging of the spinal cord, it can detect injuries to the cord that exist despite the apparent absence of bony abnormalities [45]. Nevertheless, a number of patients with SCIWORA also have no detectable lesion on MRI [46].

A common explanation for this phenomenon is transient ligamentous deformation, allowing slight spinal bony deformation, followed by spontaneous reduction. This is more often described in children who have weak paraspinal muscles, elastic spinal ligaments, and lax soft tissues, which fail to protect the spinal cord from force, but has also been described in adults. (See "Spinal cord injury without radiographic abnormality (SCIWORA) in children", section on 'Epidemiology' and "Cervical spinal column injuries in adults: Evaluation and initial management", section on 'Spinal cord injury without radiographic abnormality' and "Suspected cervical spine injury in adults: Choice of imaging", section on 'Further evaluation with magnetic resonance imaging'.)

Other possible mechanisms for SCIWORA include radiographically occult intervertebral disc herniation, epidural or intramedullary hemorrhage, fibrocartilaginous emboli from an intervertebral disc that has ruptured into the radicular artery, and traumatic aortic dissection with spinal cord infarction. MRI is invaluable for the diagnosis of these conditions.

Diagnostic evaluation of this phenomenon in the obtunded patient is discussed in detail separately. (See "Suspected cervical spine injury in adults: Choice of imaging", section on 'Further evaluation with magnetic resonance imaging'.)

MANAGEMENT

Medical care — Patients with TSCI require intensive medical care and continuous monitoring of vital signs, cardiac rhythm, arterial oxygenation, and neurologic signs in the intensive care unit [47,48]. A number of systemic as well as neurologic complications are common in the first days and weeks after TSCI, contribute substantively to prognosis, and are potentially avoidable or ameliorated with early intervention [48].

The management of medical issues specific to spinal cord injury is discussed here. The general medical care of the trauma patient is reviewed elsewhere. (See "Overview of inpatient management of the adult trauma patient".)

Cardiovascular complications — Neurogenic shock refers to hypotension, usually with bradycardia, attributed to interruption of autonomic pathways in the spinal cord causing decreased vascular resistance. Patients with TSCI may also suffer from hemodynamic shock related to blood loss and other complications. An adequate blood pressure is believed to be critical in maintaining adequate perfusion to the injured spinal cord and thereby limiting secondary ischemic injury. Albeit with few empiric supporting data, guidelines currently recommend maintaining mean arterial pressures of at least 85 to 90 mmHg and using intravenous (IV) fluids, transfusion, and pharmacologic vasopressors as needed [48-51]. Maintenance of blood pressure intraoperatively is also important. (See "Anesthesia for adults with acute spinal cord injury".)

Patients with multiple injuries often receive large amounts of IV fluids for various reasons. Excess fluids cause further cord swelling and increased damage. Therefore, fluid administration, urinary output, and electrolyte levels must be carefully monitored.

Bradycardia may require external pacing or administration of atropine. This complication usually occurs in severe, high cervical (C1 through C5) lesions in the first two weeks after TSCI [52,53].

Autonomic dysreflexia is usually a later complication of TSCI but may appear in the hospital setting, requiring acute management [54]. This phenomenon is characterized by episodic paroxysmal hypertension with headache, bradycardia, flushing, and sweating. (See "Chronic complications of spinal cord injury and disease", section on 'Autonomic dysreflexia'.)

Respiratory complications — Pulmonary complications, including respiratory failure, pulmonary edema, pneumonia, and pulmonary embolism, are the most frequent category of complications during acute hospitalization after TSCI and contribute substantively to early morbidity and mortality [36,48,55-57]. The incidence of these complications is highest with higher cervical lesions (up to 84 percent), but they are also common with thoracic lesions (65 percent).

Weakness of the diaphragm and chest wall muscles leads to impaired clearance of secretions, ineffective cough, atelectasis, and hypoventilation. (See "Respiratory physiologic changes following spinal cord injury".)

Signs of impending respiratory failure, such as increased respiratory rate, declining forced vital capacity, rising pCO₂, or falling pO₂, indicate urgent intubation and ventilation with positive pressure support [36,57,58]. Airway management may be difficult in patients with cervical spine injury because of immobilization and associated facial, head, or neck injuries. (See "Cervical spinal column injuries in adults: Evaluation and initial management", section on 'Airway management'.)

Tracheostomy is performed within 7 to 10 days, unless extubation is imminent. Patients with more severe cervical cord injuries (eg, American Spinal Injury Association [ASIA] grade A) are particularly likely to require tracheostomy [59]. The timing of tracheostomy in patients unable or unlikely to wean from ventilator support is discussed separately. (See "Tracheostomy: Rationale, indications, and contraindications".)

With a goal of preventing atelectasis and pneumonia, chest physiotherapy should be instituted as soon as possible; patients may also need frequent airway suctioning. (See "Respiratory complications in the adult patient with chronic spinal cord injury", section on 'Respiratory insufficiency' and "Respiratory complications in the adult patient with chronic spinal cord injury", section on 'Pulmonary infection'.)

Venous thromboembolism and pulmonary embolism — Deep venous thrombosis (DVT) is a common complication of TSCI, occurring in 50 to 100 percent of untreated patients, with the greatest incidence between 72 hours and 14 days [60,61]. The level and severity of TSCI does not clearly have an impact on the risk for DVT; all patients should receive prophylactic treatment. This is discussed separately. (See "Respiratory complications in the adult patient with chronic spinal cord injury", section on 'Venous thromboembolism'.)

Other medical complications

●Pain control. After spinal injuries, patients usually require pain relief. (See "Pain control in the critically ill adult patient".)

When using opiates with potential sedating properties, the need for pain control must be balanced with the need for ongoing clinical assessment, particularly in patients with concomitant head injury. Pain is often reduced by realignment and stabilization of the cervical fracture by surgery or external orthosis (see 'Decompression and stabilization' below).

●Pressure sores. Pressure sores are most common on the buttocks and heels and can develop quickly (within hours) in immobilized patients [48]. Backboards should be used only to transport patients with potentially unstable spinal injury and discontinued as soon as possible. After spinal stabilization, the patient should be turned side to side (log-rolled) every two to three hours to avoid pressure sores. Rotating beds designed for the patient with spinal cord injury should be used in the interim, if available.

●Urinary catheterization. Initially, an indwelling urinary catheter must be used to avoid bladder distension. Three or four days after injury, intermittent catheterization should be substituted, as this reduces the incidence of bladder infections [48]. Urologic evaluation with regular follow-up is recommended for all patients after TSCI [62]. (See "Chronic complications of spinal cord injury and disease", section on 'Urinary complications'.)

●Gastrointestinal stress ulceration. Patients with TSCIs, particularly those that affect the cervical cord, are at high risk for stress ulceration [63]. Prophylaxis with proton pump inhibitors is recommended upon admission for four weeks [56]. (See "Stress ulcers in the intensive care unit: Diagnosis, management, and prevention".)

●Paralytic ileus. Bowel motility may be silent for a few days to weeks after TSCI. Patients should be monitored for bowel sounds and bowel emptying, and should not ingest food or liquid until motility is restored [64].

●Temperature control. Patients with a cervical spinal cord injury may lack vasomotor control and cannot sweat below the lesion. Their temperature may vary with the environment and need to be maintained.

●Functional recovery. Occupational and physiotherapy should be started as soon as possible. Psychological counseling is also best offered to patients and relatives as early as possible.

●Nutrition. Enteral or parenteral feeding should be provided within a few days after TSCI [48].

Glucocorticoids — Methylprednisolone is the only treatment that has been suggested in clinical trials to improve neurologic outcomes in patients with acute, nonpenetrating TSCI. However, the evidence is limited, and its use is debated [65].

Efficacy — The evidence regarding the efficacy of glucocorticoids in acute TSCI is limited and, to many, unconvincing.

In animal experiments, administration of glucocorticoids after a spinal cord injury reduces edema, prevents intracellular potassium depletion, and improves neurologic recovery [19,66]. The best results were observed with administration within the first eight hours after injury [13]. Some authors believe that the main effect of methylprednisolone on the spinal cord recovery was the inhibition of lipid peroxidation, and that late administration of steroids may have little effect on lipid peroxidation and interfere with regenerative processes [67].

Two blinded, randomized controlled trials have studied the efficacy of glucocorticoid therapy in patients with acute TSCI:

●The National Acute Spinal Cord Injury Study (NASCIS) II compared methylprednisolone (30 mg/kg IV, followed by 5.4 mg/kg per hour over 23 more hours), naloxone, and placebo in 427 acute TSCI patients [67]. At one year, there was no significant difference in neurologic function among treatment groups. However, within the subset of patients treated within eight hours, those who received methylprednisolone had a modest improvement in motor recovery compared with those who received placebo. Wound infections were somewhat more common in patients who received methylprednisolone.

●NASCIS III compared three treatment groups: methylprednisolone administered for 48 hours, methylprednisolone administered for 24 hours, and tirilazad mesylate (a potent lipid peroxidation inhibitor) administered for 48 hours in patients with acute complete or incomplete TSCI [68]. All 499 patients received an initial IV bolus of 30 mg/kg methylprednisolone and were treated within eight hours of TSCI. For patients treated within three hours, there was no difference in outcomes among treatment groups at one year. For patients treated between three to eight hours, 48 hours of methylprednisolone was associated with a greater motor but not functional recovery compared with other treatments. Patients who received the longer duration infusion of methylprednisolone had more severe sepsis and severe pneumonia compared with the shorter duration of infusion; mortality was similar in all treatment groups [69].

A meta-analysis of NASCIS II with two other small trials (one positive and one negative) concluded that methylprednisolone administered within eight hours of spinal cord injury resulted in improved motor recovery [70].

Many clinicians have raised concern about complications of high-dose glucocorticoid therapy, particularly infections, in this setting. However, in the Surgical Timing in Acute Spinal Cord Injury Study (STASCIS), a nonrandomized prospective cohort study, patients who suffered a complication were less likely to have received glucocorticoid therapy on presentation than those who did not suffer a complication [71]. In addition, glucocorticoid therapy was associated with better neurologic outcomes regardless of the timing of surgical intervention.

While administration of glucocorticoids has become a standard treatment in many centers for patients with acute TSCI [13], other experts are unconvinced [72-75]. The beneficial effect of methylprednisolone compared with placebo is seen as linked to a post hoc subgroup analysis in one study (NASCIS II), although the investigators assert that early versus late treatment was an a priori hypothesis [73,76]. The cut-off times of eight hours and three hours have been seen as arbitrary. Actual gain in motor scores seen in treated patients can be interpreted as marginal. Some clinicians also believe that the potential adverse effects of glucocorticoid administration have been underemphasized, particularly with the longer, 48-hour administration [72]. Most of the recommendations for no use of steroids are based on meta-analyses of previous reports rather than on carefully controlled case series. There are, however, still neurosurgeons who administer glucocorticoids to patients with an acute TSCI in view of previous animal research, especially since so little can be done after this devastating injury.

Society guidelines and use in practice — In 2013, based upon the available evidence, the American Association of Neurological Surgeons and Congress of Neurological Surgeons stated that the use of glucocorticoids in acute spinal cord injury is not recommended [77]. Position statements from the Canadian Association of Emergency Physicians, endorsed by the American Academy of Emergency Medicine, concur that treatment with glucocorticoids is a treatment option and not a treatment standard [78-80]. A Consortium for Spinal Cord Medicine similarly concluded that "no clinical evidence exists to definitely recommend" the use of steroid therapy [81].

Use of glucocorticoids in this setting appears to be declining. In a 2006 survey of 305 neurosurgeons in the United States, 91 percent used glucocorticoids to treat patients with nonpenetrating TSCI within eight hours of injury [75]. By contrast, a 2008 survey of Canadian spine surgeons found that 76 percent did not prescribe glucocorticoids even while 76 percent had reported administering methylprednisolone five years earlier [82]. A 2013 survey of institutions in Germany and a 2014 survey of the Cervical Spine Research Society reported that a little over half of physicians use high-dose glucocorticoids for acute TSCI [83,84].

Contraindications — Methylprednisolone has been associated with increased mortality in patients with moderate to severe traumatic brain injury (TBI) and should not be administered to patients with TSCI and associated moderate to severe TBI. (See "Management of acute moderate and severe traumatic brain injury", section on 'Glucocorticoids'.)

There are few data regarding the use of methylprednisolone with penetrating injuries. However, retrospective studies suggest a higher rate of complications and no evidence of benefit [85-87]. Most clinicians do not use glucocorticoids for penetrating spinal cord injury.

Similarly, the results of NASCIS II and III studies may not apply to individuals with multisystem trauma, in whom the risk of complications is likely higher than those with isolated spinal cord injury. Patients with multisystem trauma were not specifically excluded from these trials but may have been somewhat under-represented [69].

Decompression and stabilization — There are currently no standards regarding the role, timing, and method of vertebral decompression in acute spinal cord injury [18]. Options include closed reduction using traction and open surgical procedures. Radiologic features of spinal column injuries that are associated with instability are presented in the table and are discussed separately (table 1). (See "Spinal column injuries in adults: Types, classification, and mechanisms" and "Thoracic and lumbar spinal column injury in adults: Evaluation".)

Closed reduction — For cervical spine fracture with subluxation, closed reduction methods are a treatment option. Thoracic and lumbar fractures do not respond to closed treatment methods.

This technique involves use of longitudinal traction using skull tongs or a halo headpiece. An initial weight of 5 to 15 pounds is applied; this is increased in 5-pound increments, taking lateral radiographs after each increment is applied. The more rostral the dislocation, the less weight is used, usually approximately 3 to 5 pounds per vertebral level. While weights up to 70 pounds are sometimes used, we suggest that after 35 pounds is applied, patients be observed for at least an hour with repeat cervical spine radiographs before the weight is cautiously increased further. Administration of a muscle relaxant or analgesic, such as diazepam or meperidine, may help facilitate reduction.

Closed reduction may obviate surgery and promote neurologic improvement in some cases. Early reports raised a concern that closed reduction in the setting of associated disc disruption and/or herniation has the potential to exacerbate neurologic injury [88,89]. However, more recent prospective case series and a systemic literature review suggest that this is probably not an important concern [90-92]. In one series of 82 patients with cervical subluxation injuries, early rapid closed reduction was achieved in 98 percent, failing in just two patients who required open surgical reduction [90]. The average time to achieve reduction was two hours. Disc herniation and disruption were noted in 46 percent of post-procedure magnetic resonance imaging (MRI), but these did not affect neurologic outcome.

Surgery — Goals for surgical intervention in TSCI include reduction of dislocations as well as decompression of neural elements and stabilization of the spine. There are no evidence-based guidelines regarding the indications for or timing of surgery in TSCI [93]. In general, the specific management of cervical, thoracic, and lumbar spine and spinal cord injuries depends to a large extent on a surgeon's personal experience and practice norms in his or her center.

Indications — Indications for cervical spine surgery include significant cord compression with neurologic deficits, especially those that are progressive or that are not amenable or do not respond to closed reduction, or an unstable vertebral fracture or dislocation (table 1) [94]. Neurologically intact patients are treated nonoperatively unless there is instability of the vertebral column. Most penetrating injuries require surgical exploration to ensure that there are no foreign bodies imbedded in the tissue, and also to clean the wound to prevent infection.

Defining surgical indications for closed thoracolumbar fractures has been somewhat more challenging, in part because of difficulties defining spinal instability in these lesions. The Denis anatomic-based classification based on a three-column model of spinal stability has somewhat limited clinical utility, as it does not clearly accommodate all fracture types [95]. The thoracolumbar injury severity score has been proposed as an alternative and uses a scoring system of three variables: the morphology of the injury, the integrity of the posterior ligamentous complex, and the neurologic status of the patient (table 3) [96,97]. A total score of less than four indicates a nonoperative injury; more than four, an operative injury; and four, an injury that is operative at the surgeon's discretion. This algorithm has good intrarater and interrater reliability [98]. The clinical efficacy of the algorithm itself remains to be prospectively evaluated.

Timing — The timing of surgical intervention is not defined and remains somewhat controversial [48]. Animal and some clinical studies suggest that early relief of spinal cord compression (within eight hours) leads to a better neurologic outcome [18,99-103]. However, older clinical reports suggested that early surgery led to increased medical complications and poorer neurologic outcome, perhaps as a reflection of the vulnerability of the acutely injured cord [104-106]. More contemporary studies suggest that medical complication rates are actually lower in patients who undergo early surgery, which allows for earlier mobilization and reduced length of intensive care unit and hospital stay [107-113].

One trial randomly assigned patients with complete or incomplete cervical TSCI to "early" (within 72 hours) or "late" (more than five days) surgery and found no significant difference in the improvement of ASIA grade or motor scores between the two groups [114]. A systematic review analyzed published data regarding the timing of surgical decompression and concluded that early decompression, within 72 hours, can be performed safely, without increasing systemic complications [99]. An impact on neurologic outcome of early versus late surgery could not be determined from the available data. It may be that 72 hours represents too late a cut-off time to define early surgery.

A meta-analysis that included patients from nonrandomized case series compared neurologic outcomes in 1687 patients with TSCI [115]. Those who received decompressive surgery within 24 hours had a better outcome than those treated either conservatively or with delayed surgery. An analysis of homogeneity suggested that the data in this analysis were not reliable for patients with complete TSCI. The subsequently published, nonrandomized STASCIS trial compared outcomes in those who received surgery within 24 hours (mean 14.2 hours) after injury with those whose surgery was performed later (mean 48.3 hours) [116]. After adjusting for glucocorticoid treatment and injury severity, there were 2.8-fold higher odds in improved outcomes with early surgery. Mortality and complications were similar in both patient groups.

Most clinicians consider deteriorating neurologic function after incomplete TSCI to be an indication to perform surgery as early as possible if there are no contraindications (eg, hemorrhagic shock, blood dyscrasias).

The role of early surgery with a complete TSCI (ASIA grade A) is debatable given the overall poor prognosis of these patients. While many surgeons operate to stabilize the spine, most defer the surgery to a less immediate time frame. However, many series show that a small percentage of these patients can improve, and it is possible that potential benefits for surgical decompression in this group may be maximized by earlier rather than later surgery [49].

In a 2010 survey of spine surgeons, the majority (>80 percent of 971 respondents) reported a preference to decompress the spine within 24 hours of TSCI [117]. Shorter time intervals (within 6 to 12 hours) are preferred by the majority of surgeons for certain lesions, including incomplete cervical TSCI. A 2011 report of an expert panel concurred with this approach [103].

Technical aspects — Not all surgical cases require decompression, and not all decompression cases require instrumentation and fusion. The technical aspects of the surgery are tailored to the individual case.

The anesthetic management of patients with an acute spinal cord injury is presented separately. (See "Anesthesia for adults with acute spinal cord injury".)

Investigational treatments — A number of strategies are being investigated as potential treatments of acute TSCI [16] but are not currently recommended [112]. Among others, these include:

●Spinal cord cooling [118-120]

●Electrical stimulation [121]

●Autologous macrophages [122]

●Thyrotropin-releasing hormone [123]

●Neuroprotective agents (eg, riluzole, minocycline, basic fibroblast growth factor) [124]

●Neuronal growth factors [18]

●Granulocyte colony-stimulating factor [125]

Gangliosides are endogenous compounds in cell membranes, which are believed to have the potential to protect nerve cells and promote axon growth. GM-1 ganglioside treatment has been studied in two randomized clinical trials. In one study, 34 patients were treated within 72 hours of TSCI for 18 to 32 days. Treatment was associated with a higher likelihood of improving by at least two grades compared with placebo [126]. However, in a follow-up trial of 797 patients with TSCI, there was no difference between placebo and treated patients in a similar outcome measure [127]. Gangliosides are not recommended in the treatment of TSCI [128].

PROGNOSIS — Early death rates after admission for TSCI range from 4 to 20 percent [1,4,129-132]. The patient's age, spinal cord level of injury, and neurologic grade predict survival. Severe systemic injuries, traumatic brain injury (TBI), and medical comorbidity also increase mortality [131-133]. Compared with spinal cord injuries in the thoracic cord or lower, patients with C1 to C3 injuries have a 6.6-fold increased risk of death, C4 to C5 injuries a 2.5-fold increased risk, and C6 to C8 a 1.5-fold increased risk [55]. Survivors of TSCI have a reduced life expectancy as well. (See "Chronic complications of spinal cord injury and disease", section on 'Life expectancy'.)

Rates of motor score improvements are also related to the initial severity and level of injury [134-136]. The greatest degrees of improvement are seen in those with incomplete injury and also in those without significant comorbidities or medical complications, such as infection [137,138]. Among patients with complete TSCI (American Spinal Injury Association [ASIA] grade A), 10 to 15 percent improve, 3 percent to ASIA grade D [36]; less than 10 percent will be ambulatory at one year [136]. Among patients with an initial ASIA grade B, 54 percent recover to grade C or D, and 40 percent regain some ambulatory ability. Independent ambulation is possible for 62 and 97 percent of patients with an initial ASIA grade of C and D, respectively. Most recovery in patients with incomplete TSCI takes place in the first six months [139]. The general expectations for functional recovery based on motor level are outlined in the table (table 4) [140]. These assume an uncomplicated, complete TSCI (ASIA grade A) followed by appropriate rehabilitation interventions in a healthy, motivated individual.

Patients with TSCI are at risk for a number of medical complications. These are discussed in detail separately. (See "Chronic complications of spinal cord injury and disease".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Paraplegia and quadriplegia (The Basics)")

SUMMARY AND RECOMMENDATIONS — Traumatic spinal cord injury (TSCI) is a problem that largely affects young male adults as a consequence of motor vehicle accidents, falls, or violence. (See 'Epidemiology' above.)

●Most TSCI occurs with injury to the vertebral column, producing mechanical compression or distortion of the spinal cord with secondary injuries resulting from ischemic, inflammatory, and other mechanisms. (See 'Pathophysiology' above.)

●Most TSCI is associated with injury to brain, limbs, and/or viscera, which can obscure its presentation. (See 'Clinical presentation' above.)

●The neurologic injury produced by TSCI is classified according to the spinal cord level and the severity of neurologic deficits (table 2). Half of TSCIs involve the cervical spinal cord and produce quadriparesis or quadriplegia. (See 'Clinical presentation' above.)

●The initial evaluation and management of patients with TSCI in the field and emergency department focuses on the ABCDs (airway, breathing, circulation, and disability), evaluating the extent of traumatic injuries, and immobilizing the potentially injured spinal column. (See 'Initial evaluation and treatment' above.)

●Patients with suspected TSCI because of neck pain or neurologic deficits and all trauma victims with impaired alertness or potentially distracting systemic injuries require continued immobilization until imaging studies exclude an unstable spine injury. (See 'Imaging' above.)

•All patients with potential TSCI should receive complete spinal imaging with plain radiographs or helical computed tomography (CT) scan.

•Patients with abnormal screening imaging studies or in whom TSCI remains strongly suspect despite normal screening imaging studies should have follow-up CT scanning with fine cuts through the region of interest (based on localized pain and/or neurologic signs).

•Magnetic resonance imaging (MRI) can be useful to further define the extent of TSCI and should be performed on stable patients with TSCI as well as on patients suspected to have TSCI (because of neck pain or neurologic deficits) despite a normal CT scan.

●Patients with TSCI require urgent neurosurgical consultation to manage efforts at decompression and stabilization. (See 'Decompression and stabilization' above.)

●There is limited evidence that glucocorticoid therapy improves neurologic outcomes in patients with acute TSCI, and such therapy is not endorsed by major society guidelines. (See 'Glucocorticoids' above.)

•Because the neurologic benefits are uncertain, we recommend not using glucocorticoid therapy in cases when there are clear risks associated with such therapy, such as penetrating injury, multisystem trauma, moderate to severe traumatic brain injury (TBI), and other comorbid conditions associated with risk of complications from glucocorticoid therapy (Grade 1B). (See 'Contraindications' above.)

•In other patients who present within eight hours of isolated, nonpenetrating TSCI, administration of intravenous (IV) methylprednisolone can be considered with knowledge of potential risks and uncertain benefits. The standard dose in this setting is 30 mg/kg IV bolus, followed by an infusion of 5.4 mg/kg per hour for 23 hours. (See 'Efficacy' above.)

●Patients with acute TSCI require admission to an intensive care unit for monitoring and treatment of potential acute, life-threatening complications, including cardiovascular instability and respiratory failure. Patients with TSCI should receive prophylaxis to protect against deep venous thrombosis (DVT) and pulmonary embolism (Grade 1B). (See 'Medical care' above and "Respiratory complications in the adult patient with chronic spinal cord injury", section on 'Venous thromboembolism'.)