Higher level evidence supporting spinal stabilisation is lacking

Despite spinal stabilisation being one of the most frequently performed prehospital interventions, higher grade evidence demonstrating beneficial effects is lacking [46]. From the 1970s to the 1980s the incidence and mortality of complete spinal injury decreased significantly. As this coincided with the introduction of modern spinal management strategies, authors have to varying extents credited stabilisation for this reduction [46, 60–63]. Other than these assumptions, the evidence directly supporting stabilisation consists of reports of low quality associating failure to reduce spinal mobility with neurological deterioration [6, 9, 64–67]. In contrast, in a controversial study comparing patients who had spinal stabilisation in New Mexico, USA with patients in Kuala Lumpur, Malaysia who had no spinal stabilisation, Hauswald et al. demonstrated no protective effect from stabilisation [68]. Nevertheless, taking into account the existing evidence, the anatomical perspective as well as decades of clinical experience, it is likely that the current paradigm of spinal stabilisation has played a part in the reduction of secondary neurological injury. The faculty found no reason to abandon the strategy of external spinal stabilisation. For the same reasons, after having restored the patient to an anatomical position, it is recommended that unnecessary patient handling and movement is minimised. Authors have opined that working in accordance with such a minimal handling strategy may not only reduce spinal movement but may also minimise pain as well as promote hemostasis [69].

Recommendation 3: Spinal stabilisation should never delay or preclude life-saving intervention in the critically injured trauma victim.

Spinal stabilisation may interfere with or delay life-saving intervention

The incidence of SCI in hospitalised trauma victims has been reported in the range of 0.5 to 3% [2]. Spinal stabilisation may preclude or delay the effective management of life-threatening reversible insults such as airway compromise, hypoxemia, tension pneumothorax, cardiac tamponade, haemorrhage or brain trauma which may require urgent prehospital or hospital interventions. Spinal stabilisation has been associated with difficult airway management, restricted thoracopulmonary function and delayed time to intervention [42, 44, 70–74]. In light of this, spinal stabilisation must be de-emphasised in the critically injured patient. While remaining important, spinal stabilisation should never interfere with or delay life-saving intervention nor be allowed to cause or worsen critical injury (Fig. 1).

Recognising time critical injury

Staging and defining time critical injury in trauma is controversial. Identifying patients with critical injury that are unlikely to tolerate prolonged extrication and spinal stabilisation is ultimately up to the attending clinician. Vital parameters may support the decision but should be interpreted cautiously. Nevertheless, we have opted to include a supporting definition of critical injury previously issued by The Norwegian Directorate of Health in a guideline for the management of mass casualty incidents. This definition is designed for individual triage, is based on readily obtainable clinical findings and is nationally recognised by our EMS [75].

Airway compromise, respiratory rate lower than 10 or above 30 breaths per minute, pulse frequency above 120 beats per minute, absent radial pulses or no motor response to verbal commands are signs suggestive of time critical injury. Patients with injuries designated as time critical should not suffer prolonged extrication or evacuation due to spinal concerns. Importantly, this does not imply that spinal precautions are entirely abandoned, but only applied to an extent and in a way that does not delay extrication nor intervention.

The lateral trauma position

Historically, first responders without advanced airway skills have placed unconscious or obtunded victims in the recovery position in order to facilitate the clearance of fluids and to maintain airway patency [45]. Unfortunately, this contradicts the principle of spinal stabilisation in trauma victims as it generates unacceptable spinal movement [76]. The lateral trauma position (LTP) is a variation of the established recovery position that is achieved using a modified two-person log-roll with manual cervical spine control and, eventually, blankets and a rigid collar for cervical stabilisation [77]. While the log-roll involved may generate spinal motion, this may be offset by the beneficial effects of gaining airway patency and clearance. Clinicians not trained in advanced airway management should be encouraged to consider the LTP when transporting obtunded patients.

Recommendation 4: Victims of isolated penetrating injury should not be immobilised.

Triaging tools

In order to address over-triage, authors have advocated implementing triaging tools to assist in identifying low-risk patients who do not require stabilisation [41, 78–81]. The National Emergency X-radiography Utilisation Study (NEXUS) tool and the Canadian C-Spine Rule Criteria (CCR) were originally developed to aid physicians in determining which trauma patients require imaging of the cervical spine [2, 82, 83]. Protocols similar to NEXUS have proven to be useful triaging tools for prehospital spinal stabilisation. Since the early 1990s the Fresno/Kings/Madera EMS system in California have implemented a selective stabilisation protocol similar to NEXUS. In a 2001 retrospective review, Stroh and Braude reported that this protocol had a 99% sensitivity for the correct stabilisation of patients with actual cervical injury [84]. In a prospective observational study of EMS personnel in Maine, also using a tool similar to NEXUS, Burton et al. found that the protocol sensitivity for stabilisation of any spinal fracture was 87% with a negative predictive value of 99.9% [85].

Triaging tools based on clinical findings reduce over-triage

Authors have recommended implementing tools that, similar to NEXUS, are predominantly based on clinical findings [78, 86, 87]. Tools that emphasise the mechanism of injury result in over-triage without increasing accuracy. In a prospective review of 498 trauma patients, Hong et al. found that 95.4% of patients would have been immobilised if EMS personnel had stabilised in accordance with the mechanism based 7th edition PHTLS criteria. In contrast, stabilisation in accordance with protocols based on clinical findings, NEXUS or Hankins protocols, would result in stabilisation rates of 68.7% and 81.5%, respectively. All patients with actual spinal injury would have been stabilised using any of the protocols [88]. In 1999, Muhr et al. reported how the implementation of an out-of-hospital clearance protocol based on clinical findings reduced stabilisation by one third [87]. These and other reports provide evidence of how EMS successfully can implement selective prehospital stabilisation strategies given that they are coupled with training and clinical governance [89–93]. The faculty recommends applying the NEXUS criteria on the entire spine for triage in the prehospital setting. In the absence of midline tenderness, focal neurologic deficit, altered level of consciousness, intoxication, and significant distracting injury, it is safe to withhold stabilisation.

Recommendation 6: Cervical stabilisation may be achieved using manual in-line stabilisation, head-blocks, a rigid collar or combinations thereof.

The efficacy and harms of the rigid cervical collar

No high quality studies have identified the true efficacy of the rigid collar. The existing evidence is difficult to compare due to variations in methodology and types of collars tested [94]. However, numerous studies document how the application of a rigid cervical collar will limit motion in the cervical spine [34, 36, 95–101]. It is also apparent, from these same studies, that movement restriction is limited. Moreover, there is a growing body of evidence documenting harm. As rigid collars achieve cervical stabilisation through compression of the mandible, mouth opening will be reduced. Thus, application may impede breathing and airway management including the clearing of vomit or secretions [43, 102]. Rigid cervical collars can increase intracranial pressure by inducing pain or through blocking cranial venous return [103–105]. In a study on cadavers with an artificially induced unstable C1-C2 lesion, Ben-Galim et al. demonstrated how cervical traction from a collar caused separation between C1 and C2, suggesting a mechanism that could aggravate injury [106]. Severe neurological deterioration has been reported in patients with ankylosing spondylitis after receiving triple stabilisation [107, 108]. Finally, rigid collars may induce pain or discomfort that may trigger non-compliance, agitation and even increased spinal movement in some patients [109–111]. Cervical collar use has also been associated with pressure point ulceration, necrosis and mandibular nerve palsy with prolonged use [112–116].

The rigid collar should not be applied routinely

The aforementioned reports support a selective approach to rigid collar use. While collars are safe to use in the majority of patients, they should be used selectively in patients with traumatic brain injury, airway compromise, ankylosing spondylitis or agitation. In such cases the collar may be withheld or used intermittently. The collar may provide support during certain manoeuvres, such as in stretcher transfers or during evacuation from a vehicle, after which the collar may be opened or removed [99, 117]. With adequate MILS this can be achieved with minimal spinal displacement [118]. Transport may proceed using only MILS and/or head blocks. Holla et al. recently demonstrated how the addition of a rigid collar did not result in improved movement restriction in volunteers already strapped to a rigid stretcher with head blocks [102]. Patients with a kyphotic spine, such as in ankylosing spondylitis, should be stabilised in a position similar to their habitual spinal curvature [108].

Recommendation 7: Transfer from the ground or between stretchers systems should be achieved using a scoop stretcher.

The log-roll may generate undue spinal motion and should be avoided in favour of alternative techniques

Log-rolling has traditionally been used to transfer the patient onto or off stretcher systems or to gain access to patients back for examination, despite authors questioning its safety [119]. The log-roll is a potentially dangerous procedure as it may cause fracture dislocation, pain, distress or clot disruption in patients with pelvic fractures or other injuries. The diagnostic value is limited [69, 120, 121]. Moreover, as the head, hips and pelvis are of different diameters, spinal motion is inherent to the technique, and several studies have demonstrated how log-rolling generates more motion than readily available alternative techniques such as lift-and-slide or scoop stretcher techniques [122–128]. The faculty believes that the potential spinal motion generated by the log-roll may be further aggravated in the prehospital setting where it is commonly performed with limited personell and under difficult working conditions. Usage of the technique in the prehospital context should therefore be minimised, if not abolished.

For transfers from the ground or between stretcher systems, we recommend employing a scoop stretcher system. As the scoop stretcher is split vertically and then reassembled underneath the patient, transfer from the ground or between stretchers requires minimal or no rolling [129]. Stabilisation and comfort has been demonstrated to be comparable or better than that of the classic backboard [122, 127, 130].

Recommendation 8: Patients with potential spinal injury should be transported strapped supine on a vacuum mattress or on an ambulance stretcher system.

Recommendation 9: Hard surface stretcher systems may be used for transports of shorter duration only.

Hard surface stretcher systems

The backboard was designed to facilitate extrication but has since its inception been used as a transportation device and quickly became the gold standard for spinal stabilisation during transport [131, 132]. The literature, on the contrary, suggests that it is not appropriate for transports of longer duration. Within short time, patients will develop significant discomfort and moderate to severe pain [133–135]. Prolonged exposure may result in pressure ulcers [136, 137]. Pain and discomfort may also result in undue voluntary spinal movement [133]. The scoop stretcher, like the backboard, has hard surfaces that could induce pain, discomfort or pressure point injury. While it is an excellent extrication device and an appropriate transportation device for short distances, for longer duration transport the scoop stretcher, like the backboard, should be removed after transferring the victim onto a vacuum splint mattress or onto a standard ambulance trolley.

Soft surface stretcher systems

The vacuum mattress, while not rigid enough for extrication, is a useful transportation device. As vacuum is applied, the mattress moulds to the patient’s contours, minimising point pressure, making it more comfortable, less painful and, arguably, less likely to produce ulceration [73, 138–141]. The vacuum mattress has been shown to provide a similar, or superior, degree of stabilisation when compared to that of the backboard [132, 139, 140, 142].

Recommendation 10: Patients should under some circumstances be invited to self-extricate from vehicles.

Self-extrication

Over the years, authors have argued that this practice often is unnecessary, resulting in prolonged extrication times and avoidable complications related to spinal stabilisation. Authors have argued that spinal movement within the normal range of motion requires so little energy, of many magnitudes less than the energy at the initial impact, that it is highly unlikely to cause further injury. Furthermore, the alert victim’s own muscular tone will suffice to protect the spine from further injury [41, 68, 144]. In 2013, the British Faculty of Pre-Hospital care acknowledged this in a statement recommending that the fully alert patient a potential spinal injury who is without distracting injury, should be allowed to self-extricate without external stabilisation [2]. Unfortunately, such a position is supported by very few studies. Shafer and Naunheim, in 2009, demonstrated how volunteers stabilised only with a rigid collar who exited a vehicle on their own volition, generated less spinal motion than when extricated using traditional assisted longboard techniques [145]. More recently, Dixon et al., in a biomechanical study on healthy volunteers found that controlled self-extrication without collar generated less movement in the cervical spine when compared to equipment aided extrication techniques [146].

A generous approach to self-extrication

Despite the scant evidence, we recommend self-extrication in some circumstances. As long as patients with back or neck pain are not obtunded, not under the influence of any drug, and without significant distracting injury, they should be invited to self-extricate to a nearby stretcher system. The prerequisite for self-extrication is that it is done under safe conditions. Should there be concerns about safety, then strategy defaults back to traditional extrication techniques. The patients should, after lying down on a stretcher system, have full external stabilisation for final evacuation and transport as they may then be subject to external force that may overwhelm their muscular protection.