The current recommendation in reference to the immobilization of trauma patients is to apply selective immobilization, identifying those individuals that may benefit from this measure.5–7 The devices used for immobilization are associated to complications such as increased intracranial pressure, difficulties accessing the airway, restricted lung function, pain, agitation, pressure ulcers and the prolongation of patient transfer time. The use of a cervical collar does not completely limit mobility of the cervical spine. When vertebral damage is suspected, the entire spinal column is to be immobilized, since the presence of additional non-contiguous vertebral injury occurs in up to 20% of the cases.8 Although strong supporting evidence is lacking9,10 and negative effects may even occur, the recommendation to immobilize patients with suspected vertebral-spinal cord injury is based on anatomical, mechanical and clinical considerations—the aim being to prevent the development or worsening of SI in the presence of unstable vertebral damage.11

Immobilization is not indicated in patients with penetrating trauma in the absence of neurological symptoms.6,7,12 Algorithms have been developed in the case of closed trauma patients, in an attempt to avoid indiscriminate immobilization.13–16 The most widely used criteria in the out-hospital emergency care setting are those proposed by the National Emergency X-Radiography Utilization Study (NEXUS)14 and the Canadian C-Spine Rule (CCSR).15 The NEXUS establishes 5 low risk criteria which when met could discard cervical damage (Table 2). The CCSR combines high and low risk criteria and the capacity to rotate the head 45°. In this regard, the patient is without risk and therefore will not need immobilization if one low risk criterion is met and the patient can voluntarily rotate the head 45°17 (Table 3). These or similar immobilization algorithms have been incorporated to emergency team intervention protocols throughout the world.5,17–21 The National Institute for Health and Care Excellence (NICE) guide on the assessment and management of spinal injuries favors the CCSR criteria in evaluating the cervical spine, though its immobilization recommendations combine NEXUS and CCSR criteria. Although it is agreed that a selective immobilization algorithm should be applied, there is no consensus regarding which algorithm is best. In the United Kingdom, the Faculty of Pre-Hospital Care drafted a consensus document suggesting that an algorithm similar to that proposed by the NEXUS might be appropriate, though this remains to be confirmed.6 There is controversy in the literature regarding inclusion of the injury mechanism as a risk criterion; whether different algorithms should be used in conscious and unconscious patients; and regarding which patients are amenable to auto-rescue measures.6,22,23

Table 3.

Canadian C-Spine immobilization criteria.

The usual immobilization technique involves the use of a spinal table with straps and head fixation, with the use of a cervical collar.7,24 However, this practice—a standard for many emergency care teams—is changing. The immobilization methods may be different during rescue, retrieval and transport, and can be modified according to the clinical condition of the patient, placing priority on the ABC approach. In the management of trauma patients, spinal protection maneuvering should be the norm until clinical evaluation can be made. Immobilization devices may be difficult to fit in patients with certain characteristics (short neck or spinal deformities) or in individuals that fail to cooperate or are restless, and fitting moreover takes time. Under these circumstances, and when “rapid rescue” is needed, manual immobilization can be maintained,6,21 followed by fitting of a cervical collar of adequate size, if indicated. Mobilization is to be carried out trying to preserve the vertebral axis, applying linear traction upon the cervical spine and avoiding deformities of the spine and movements of the head and neck. The High Arm In Endangered Spine (HAINES) or modified HAINES technique is currently more widely recommended than the classical log-roll procedure, since it causes less mobilization of the spine.25 The long spinal table, Kendricks rescue device and scissor-type stretcher can also be used during rescue. The National Association of EMS Physicians and the American College of Surgeons Committee on Trauma limit use of the spinal table to certain situations,23 regarding it as a “rescue” device, rather than a transport device.6 For transport purposes, the vacuum mattress combined with a cervical collar affords similar or better immobilization compared with the spinal table, and is more comfortable.21,26 Transport can also be safely made using the ambulance stretcher, correctly fastening the patient with straps together with the cervical collar.23 In children, the disproportion between the head and rest of the body may require the addition of cushioning in order to preserve body posture and avoid neurological deterioration. The same applies to individuals with spinal kyphosis. Mobilization and transfer should be limited as far as possible. Inter-exchanging immobilization means are therefore needed between the pre-hospital and in-hospital settings.

The suspicion or confirmation of SI, together with safe transfer to hospital, constitutes the first step in correct patient management. In the event of life-threatening situations, the patient should be taken to the nearest hospital center, followed by referral to a center specialized in SI as soon as possible (preferably within the first 24h). The choice of transport will depend on the available means, patient stability and the distance to the hospital center.27 Transport by air does not necessarily have to be used, though it may be recommended for distances of over 80km.28

Continuous monitoring of respiratory failure is required in all patients with cervical SI.29 Effective respiratory muscle function is lacking in the case of C1–C2 injuries. Lesions at C3–C4 level in turn produce bilateral phrenic nerve paralysis, and ventilation becomes dependent upon the accessory muscles. Consequently, patients with complete motor damage above C5 almost invariably require ventilatory support. Tetraplegic patients with adequate ventilation at the expense of important respiratory labor must be intubated and placed under mechanical ventilation without delay. Injuries below C5 cause paralysis of the intercostal and abdominal muscles, and most affected individuals will require ventilatory support.

The loss of respiratory capacity results from exhaustion of the ventilatory muscles, ascending spinal cord edema or bleeding, the accumulation of secretions, atelectasis, other associated traumatisms, or other patient conditions. Lesions below T5 usually do not produce respiratory failure of neuromuscular origin.

Lung edema secondary to traumatic SI may be cardiogenic (due to high catecholamine and beta-endorphin levels) and/or non-cardiogenic (pulmonary hyperpermeability). Echocardiographic studies and the new cardiopulmonary assessment devices may be of help in determining the predominant mechanism. Pulmonary thromboembolism and pneumonia are complications that tend to manifest after the first 24h.

Endotracheal intubation may be particularly difficult in patients with cervical SI. Intubation is sometimes needed before the lesion and is location are confirmed. Consequently, patients requiring to be intubated following trauma are to be managed as cervical SI cases, ensuring preservation of the airway with as little movement of the cervical spine as possible. Emergency intubation in the case of suspected or confirmed cervical SI must be carried out with a rapid induction sequence (to reduce coughing and spontaneous movements), with linear manual traction stabilization of the spine.24 Traction must be done carefully in order to avoid distraction—particularly in lesions affecting the occipitocervical junction. In patients with cervical immobilization, use of the Airtraq device reduces the risk of failure.30 There is a lack of evidence on the usefulness of other intubation devices.

If securing the airway and ventilatory support are not initially required, the need for intubation should be monitored by means of PCO2 and spirometry with the measurement of vital capacity (which shows excellent correlation to other lung function tests) and peak inspiratory pressure (PImax).31,32 This offers a simple and global evaluation of inspiratory muscle strength, and constitutes one of the best point of care markers. In general terms, PImax estimates the strength of the inspiratory muscles (diaphragm), being recorded under residual volume and with maximum inspiratory effort. Ventilatory failure most often occurs beyond the fourth day after trauma. This is relevant for monitoring and for the decision to maintain postoperative intubation in patients undergoing early surgery – particularly those subjected to neck surgery.

Adequate and early resuscitation with fluids is a priority concern in the treatment of hypotension, in order to maintain adequate perfusion and avoid secondary damage to the central nervous system.33 The most appropriate resuscitation target and optimum mean blood pressure (MBP) value for maintaining spinal cord perfusion are not clear. Non-controlled studies using fluids and vasopressor drugs to secure an MBP of 85mmHg during a minimum of 7 days in patients with acute SI have reported favorable results.34–36 Many studies underscore that hypotension must be identified, seeking the cause and starting resuscitation with fluids. Further studies are needed to define the ideal MBP and the role of fluids and drug support in this regard.

The initial base defect or lactate concentration can be used to assess shock and the need for resuscitation with fluids.37

Neurogenic shock is common in patients with acute SI above level T6.38 However, before assuming that SI is the origin of hypotension, other causes of hemodynamic instability in trauma patients should be discarded, such as bleeding, pneumothorax, myocardial damage, cardiac tamponade, sepsis related to abdominal injuries, adrenal gland failure, etc. The physical examination and symptoms are scantly sensitive in patients with SI; thoracoabdominopelvic imaging studies are therefore required in order to rule out other possible causes of hypotension.

Neurogenic shock is secondary to sympathetic denervation, which leads to arterial vasodilatation and blood sequestration in the venous compartment, together with interruption of cardiac sympathetic innervation (T1–T4) with unopposed vagal nerve activity that promotes bradycardia and reduces cardiac contractility. Neurogenic shock is therefore characterized by a drop in arterial pressure and systemic vascular resistances, with a variable heart rate response. Spinal injury above T6 should be suspected in all polytraumatized patients with hypotension and bradycardia, particularly in the absence of signs of peripheral vasoconstriction.

Experimental studies have shown that hypotension and hypovolemia are deleterious for the damaged spinal cord, contributing to hypoperfusion and perpetuating the secondary damage. Higher lesion levels are correlated to more severe hypotension. It is essential to restore intravascular volume and, if necessary, start vasopressor drug treatment.36 Bradycardia-dysrhythmias can lead to hypotension and asystolia – this being more frequent in the first two weeks after injury. Hypoxia, laryngeal or tracheal manipulation maneuvers, and hypothermia, worsen bradycardia. Atropine must be at hand whenever airway manipulation is intended in patients with cervical SI. The need for chronotropic drugs (atropine, adrenalin, noradrenalin) or phosphodiesterase inhibitors (aminophylline, theophylline) is more frequent in patients with cervical SI. The β2 receptors are an attractive target for modulating vagal nerve activity and thus raising the heart rate in the absence of significant risks.39 A cardiac pacemaker to deal with bradycardia is reserved for those patients with symptomatic bradycardia that fails to respond to drug treatment.

The autonomic nervous system is interrupted in cases of injury above T6, giving rise to alterations in thermoregulation due to the incapacity of the hypothalamus to control temperature due to the loss of vasomotor control. The affected patients can present hypothermia as well as incapacity to dissipate body heat.40 Temperature monitoring is essential during management of the acute phase.

The immediate effects of complete SI include loss of movement and sensitivity below the level of the lesion. Furthermore, the patient may suffer neurogenic shock, flaccid paralysis of the bladder and intestine with urinary retention and paralytic ileus, and alterations of all systems below the level of the lesion. Initial phase spinal shock is also observed, characterized by a loss of reflex activity and flaccidity below the level of the lesion. This period usually lasts days or weeks, and is subsequently followed by spasticity.

The clinical diagnosis starts with the basic neurological examination. Loss of consciousness and the mechanism of injury are to be evaluated as far as possible. If the patient is conscious, detailed motor and sensory assessment is indicated. However, such evaluation is often not possible in polytraumatized patients, which moreover may be sedated or intubated. In these circumstances, it is therefore advisable to manage the patient as if vertebral-spinal cord injury were effectively present. Table 1 describes the alarm signs that may cause us to suspect the presence of such damage at initial exploration.

The neurological exploration is made according to the International Standards for the Neurological Classification of Spinal Cord Injury (http://asia-spinalinjury.org/wp-content/uploads/2016/02/International_Stds_Diagram_Worksheet.pdf) of the American Spinal Injury Association (ASIA) and the International Spinal Cord Society.41 This system describes the level and extent of the lesion, based on a systematic evaluation of sensory and motor function. In addition to being of prognostic value, this assessment based on the ASIA scale is of use in monitoring the patient neurological course. This evaluation has been protocolized to be carried out upon patient admission, after 72h, after one month and at discharge, and also in certain situations in which worsening of SI may be suspected. Table 4 offers basic definitions of terms used in assessing SI.

In relation to the sensory exploration, we assess 28 dermatomes in each side of the body (C2 to S4–S5, considering the latter as a single dermatome). For each dermatome, we assess sensitivity to pain (needle prick) and superficial tactile sensitivity (cotton friction) based on a three-point scale (0=none; 1=impaired; 2=normal). Deep anal sensitivity is also evaluated by rectal digital testing, registered as either present or absent.

We likewise explore muscle balance in 5 key muscle groups of the upper extremity and in 5 of the lower extremity (in each side of the body). Muscle balance is scored from 0 to 5 according to the Medical Research Council scale. Elbow flexion (C5), carpal extension (C6), elbow extension (C7) and flexion of the fingers (C8) and abductors of the little finger (T1) are evaluated in the upper extremity. Hip flexion (L2), knee extension (L3), ankle dorsiflexion (L4), extension of the big toe (L5) and ankle plantar flexion (S1) are evaluated in the lower extremity. As part of the motor evaluation, we assess voluntary contraction of the external anal sphincter through rectal digital testing, registered as either present or absent.

The sensory level is the most caudad intact dermatome referred to both sensitivities (tact and pain). The motor level in turn is defined by the most caudad key muscle with a muscle score of at least 3, provided the key muscles above that level are intact (muscle score 5). In the regions where there are no dermatomes for evaluation, the motor level is taken to be the same as the sensory level.

The neurological level of the lesion refers to the most caudad segment of the spinal cord in which both motor and sensory function are bilaterally normal. It is the most cephalic of the sensory and motor levels.

Spinal cord injury is generally classified as “complete” or “incomplete”, based on the presence of sacral preservation. The latter refers to the presence of sensory or motor function in the most caudad sacral segments (preservation of light tact or pain sensation in S4–S5, deep anal sensation, or voluntary contraction of the anal sphincter).

A complete lesion is defined as the absence of sacral preservation (no sensory or motor function in segments S4–S5), while an incomplete lesion is defined as the presence of sacral preservation (some preserved sensory or motor function in segments S4–S5).

The ASIA scale comprises 5 grades (Table 4). Grade A corresponds to complete SI; grades B, C and D are incomplete lesions of different grades; and grade E is indicative of normal motor and sensory function (this grade being applied provided the patient has suffered some degree of spinal cord injury that has recovered).

• •

  Central spinal cord. Central spinal cord injury is the most frequent type of incomplete lesion. It is characterized by greater motor weakness in the upper extremities than in the lower extremities, sphincter alterations and variable degree of sensory involvement. These presentations are more common in patients with previous degenerative changes of the vertebras that suffer hyperextension.42

• •

  Brown-Sèquard (cord hemisection). This presentation is characterized by loss of deep sensitivity ipsilateral to the lesion, with contralateral alteration of heat-pain sensitivity. Pure cord hemisection is rare in traumatic SI, combining Brown-Sèquard symptoms and central cord symptoms (Brown-Sèquard plus syndrome).

• •

  Anterior spinal cord. This presentation can result from direct damage to the anterior part of the spinal cord due to retropulsion of a bone or disc fragment, or can be caused by damage to the anterior spinal artery. It is characterized by paralysis and alteration of heat-pain sensitivity, with preservation of the posterior cords (light tact sensation and positional sensitivity).

• •

  Medullary cone. These are lesions of the sacral spinal cord (cone) and lumbar nerve roots within the neural canal. Medullary cone injuries result in paralysis of the bladder, intestine and lower extremities. Depending on the affected level, these cases can manifest as mixed superior and inferior motor neuron syndromes.

• •

  Cauda equina. These are lesions of the lumbosacral roots below the medullary cone, located within the neural canal. The clinical manifestations are similar to those of medullary cone injury, making it sometimes difficult to distinguish between the two presentations.

The indications of radiological exploration in diagnosing traumatic SI depend on the circumstances of trauma, the need to perform tests for diagnosing associated lesions, and the availability of techniques in each individual center.

Clinical protocols contemplated in the Eastern Association of Trauma39 guides allow us to discard cervical spine injuries in asymptomatic patients without performing imaging explorations. The NEXUS14 and CCSR criteria15 are very sensitive in discarding significant cervical injuries without conducting radiological studies. Patients with spinal pain, neurological symptoms, and unconscious trauma patients, require radiological evaluation. In patients with masking associated injuries or short-duration altered consciousness, we can decide to use spinal protection measures until the mentioned injuries are treated or sufficient consciousness is regained to allow adequate clinical assessment. However, if urgency is a priority concern, these individuals must be evaluated under the same criteria as in unconscious polytraumatized patients.43

In patients requiring a radiological study, it is advisable to evaluate the entire spine, since the incidence of multiple vertebral fractures is close to 20%.8 Plain X-rays with anteroposterior and lateral projections of the entire spine were traditionally indicated, together with transoral projection in the cervical spine, though the current tendency is to perform computed axial tomography (CAT) with sagittal and coronal reconstructions.44 Spiral CAT, while more expensive, is more sensitive and specific than conventional radiology44,45: it allows easier identification of fractures that might go undetected on plain X-rays; ensures correct visualization of the occipitocervical and cervicothoracic junctions; affords full visualization of the entire spine with reconstructions allowing improved characterization of canal invasion; and provides information of help in surgical treatment. Plain X-rays studies in flexion and extension are useful for assessing instability secondary to ligamentous damage, but should be avoided in the acute phase of patients with neurological deficits.

Emergency magnetic resonance imaging (MRI) is usually difficult to perform in acute polytraumatized patients. Emergency MRI is indicated in traumatic SI in those cases where the neurological defects cannot be explained by the radiological findings, or when neurological deterioration occurs, with a view to discarding the possible presence of epidural hematoma or other causes indicating emergency surgical treatment. Magnetic resonance imaging should be performed as soon as allowed by the clinical condition of the patient, because it is able to detect damage to soft tissues and ligaments in the setting of neurological damage, and is moreover able to adequately characterize the type of SI involved (contusion, edema, bleeding, cord sectioning). This information in turn can be complemented by clinical evaluation based on the ASIA scale in order to establish the prognosis of SI. In patients with cord sectioning or intramedullary hemorrhage, neurological recovery generally can be expected to be poor or inexistent, while in individuals with spinal cord edema or contusions, greater recovery may be observed.46,47 Furthermore, MRI allows us to evaluate canal involvement, the degree of spinal cord compression, and the length of the lesion – these factors also being important for predicting the severity of the damage and patient recovery.48,49