For the past few decades, the importance of spinal immobilisation has been emphasised in standard texts, trauma education such as the international Advanced Trauma Life Support (ATLS) & Pre-Hospital Trauma Life Support (PHTLS) courses and in many emergency medical service (EMS) operating procedures. Such approaches often mandate the universal application of spinal immobilisation to trauma patients [1,2,3].
This approach is based on the logic that spinal injuries are not uncommon in trauma patients and that EMS must adhere to the notion of primum non nocere (‘first do no harm’), by ensuring that potential spinal injuries are not exacerbated during patient extrication and transfer.
As a result, traditional dogma has emphasised the application of a semi-rigid cervical collar (C-collar) for all trauma patients and has until relatively recently included the mantra of transfer of patients on a rigid ‘spinal’ board.
“I’ve been teaching the ‘ATLS handshake’ on ATLS-EMST courses for the past few years…but have become increasingly concerned with the blind application of semi-rigid collars to patients, without considering nuance of the circumstances.
“This is especially relevant in rural areas, where facilities for imaging are often limited – the requires subsequent transfer if clinical suspicion exists and the possibility of patients suffering harm from being immobilised in C collars for prolonged periods”
The rationale for spinal immobilisation is based upon the principles that :
- trauma patients are at risk of spinal injuries
- trauma patients may have an unstable injury of the spine
- further movement of the spine could cause further damage than that originally sustained
- spinal injures can cause significant morbidity and mortality
- the application of a semi-rigid C-collar and spinal immobilisation device prevents potentially harmful movement of the spine
- measures to provide immobilisation of the spine are in themselves harmless and can be applied to a large number of patients as a routine ‘precaution’
Spinal injuries are associated with trauma, of that there is no doubt. Global incidences of spinal cord injury are reported to vary from 8-246 cases per million, likely reflecting reporting and coding variability as well as socioeconomic and geographic differences . Similarly the sequelae of spinal cord injury are not contested – the potential adverse morbidity and mortality from cervical injuries can be catastrophic, as evidenced by those unfortunate enough to have suffered traumatic injury to the cervical spine (C-spine) with neurological sequelae. High cervical injuries can be fatal or cause quadriplegia, lower spinal injuries can cause permanent disability, reduced ability to work and decreased quality of life, with deficit relating to level of spinal cord injury. However, the times they are a’changing.
The Dogma of Cervical Collar Application
The notion of minimising further movement of the cervical spine in a trauma patient appears intuitively logical, especially mindful of the potential catastrophic damage of a cervical injury. Thus traditional teaching and practice has been for early cervical spine immobilisation. This may take the form of manual immobilisation, application of a semi-rigid C-collar or the use of head blocks and tape.
Whilst manual immobilisation may be appropriate for initial responders prior to the application of a collar, it does require both hands of the rescuer. Thus use of C-collars has been practiced widely by many prehospital services to both free up personnel from the provision of manual immobilisation and in accordance with the notion of minimising further injury to the C-spine.
But why worry about the use of C-collars? Are they not a simple intervention where the potential benefits mandate routine application? Sadly the answer is no. A more nuanced and targeted approach is necessary. This is in recognition of not only little evidence of benefit, but also the potential for harm [5,6]. Poorly fitted collars are not uncommon and may cause harm.
Other harms may include:
- obscuring injuries by presence of a collar 
- increasing scene time [6,7]
- raising intracranial pressure by impediment of venous return [8,9,10]
- limiting mouth opening and hampering airway management [11,12,13,14,15]
- causing lateral flexion or unrestricted movement if incorrectly applied 
- forcing those with pre-existing deformity or fragility of the C-spine into harmful positions via injudicious application 
- pressure necrosis from the collar itself if left in situ for > 2 hours [18,19]
Trauma patients are a heterogeneous population. It is perhaps easiest to consider mechanism of trauma as being classified into either penetrating or blunt to guide further discussion. These classifications are not exclusive; some patients may have a combination of blunt and penetrating injuries.
The literature shows that routine application of C-collar to patients with isolated penetrating trauma is at best of no benefit and indeed may cause harm . Oteir et al have recently conducted a systematic review confirming that C-collar application in penetrating trauma is associated with increased scene times and concealment of neck injury, with C-collar application associated with increased risk of mortality . The overall recommendation is that patients with a single isolated penetrating injury do not need a C-collar, unless there is additional evidence of significant blunt force trauma or flexion/extension injury to the C-spine.
What then of the remaining patients? Those with blunt force trauma and a significant mechanism of injury, or those at particular risk (such as the elderly)? Or the unconscious patient? Although controversial this offers an opportunity to re-examine established dogma with a view to developing a more nuanced practice.
(i) the awake patient with blunt trama:
One can posit that application of a C-collar will minimise movement and reduce the risk of downstream injury. However the application of a semi-rigid C-collar is not proven to minimise spinal movement. A 2015 International Liaison Committee on Resuscitation (ILCOR) draft consensus statement analysed the question as to whether spinal motion restriction affected key ‘critical outcome’ measures (including changes in neurological injury, complications, overall mortality, pain, patient comfort, movement of the spine and overall outcome . The reviewers note the paucity of evidence for restriction of spinal motion, with C-collars – commenting that studies are limited as based on either healthy volunteers and cadavers. The consensus statement argues against routine spinal motion restriction for both adults and children with blunt trauma and suspected C-spine injury. Tellingly this draft states “because of proven adverse effects in studies with injured patients, and evidence concerning a decrease in head movement only comes from studies with cadavers or healthy volunteers, benefits do not outweigh harms, and routine application of cervical collars is not recommended”
Hauswald and colleagues argued back in 1998 that the initial traumatic impact will cause injury to the spinal cord with subsequent movement unlikely to cause further damage . Subsequent to this, a 2001 Cochrane review failed to demonstrate a benefit to routine application of cervical collars ; yet a decade later this has remained a ‘standard of care’ espoused by many emergency services.
Perhaps most problematic is the otherwise well patient who complains of midline cervical spine tenderness in the absence of a neurological deficit. Such patients have traditionally been placed in full spinal precautions and transferred for imaging – indeed, examples abound of patients involved in motor vehicle collisions who have self-extricated and been placed in an undamaged vehicle pending assessment, who have then developed cervical pain and required extrication with full spinal precautions – even to the point of being cut from a vehicle that was not involved in the accident.
In the UK, unthinking application of a spinal immobilisation protocol has lead to farce after patients self-extricate from crashed vehicle…relocated to police car…then complained of neck pain!
This begs the question as to the natural history of patients who may have suffered a C-spine injury but are neurologically intact and otherwise well. How should EMS move forward in the face of uncertainty of the benefits of routine C-collar application and the understandable concern of an unstable C-spine injury causing devastating neurological sequelae?
For conscious trauma patients, the development of reliable cervical spine screening tools such as the Canadian C-spine rules and the NEXUS criteria allows the possibility of rapid exclusion of clinically significant C-spine injury according to defined criteria . However this requires consistent application of these rules, which may be problematic where significant heterogeneity exists in EMS responders (for example, different skills mix between volunteer ambulance officers, paramedics, intensive care and extended care paramedics). Nevertheless, whilst the superiority of NEXUS vs Canadian C-spine rules has been contested , they can be taught relatively easily. However as Benger & Blackham comment “the specificity of both rules is such that they still mandate immobilisation in a large number of injured patients” .
A more pragmatic approach may be to consider that those patients who are able to self-extricate and mobilise are often able to stabilise and protect their own C-spine . Thus a mature prehospital service may decide not to routinely apply C-collars to patients with penetrating trauma alone, to those who fulfil validated screening tool criteria and conscious, stable and cooperative patients without distracting injury who are able to protect their own C-spine. Some patients may wish to lay supine or find comfort in the support of a collar and this should be accommodated where possible. However for the vast majority of patients, a C-collar merely indicates that a potential cervical injury has occurred (one could equally well use a red dot or triage tag). Hence Benger & Blackham’s suggestion of transporting such patients in the “position of comfort” and ensuring that any concern regarding potential spinal injury is communicated to the receiving team .
(ii) the patient with reduced level of consciousness and blunt trauma
As discussed above, conscious patients, even those with a suspected cervical spine injury, are able to protect their own C-spines . However those patients with a reduced level of consciousness cannot. Nor may they be able to protect their own airway. A balance needs to be determined between airway management and spinal immoblisation. On occasions there may be competing priorities between the need to minimise spinal movement AND maintain airway patency . Use of a left lateral or HAINES (High Arm In Endangered Spine) position may be required to provide a stable airway whilst minimising the risk of secondary spinal injury, with no reported difference between techniques .
However both of these positions did produce potentially clinically significant movement of the spine; moreover it must be remembered that blunt trauma patients may also have other significant injuries, not least pelvic disruption. A strong argument can therefore be made for early airway intervention in such patients, then immobilisation in a manner to minimise further handling and protect not just spine but also pelvis and long bones, minimising further injury and the risk of decubitus ulceration. This will be discussed in greater detail under consideration of total spinal immobilisation below.
For the unconscious patient, some form of C-spine immobilisation may be considered – but not at the expense of impeding airway management. Other considerations such as the effect of C-collar on intracranial pressure (ICP) and scene time must also be considered. In a provocative paper, Sundstrom et al. propose abandoning the use of C-collars in this cohort, transporting those with unsecured airways as expeditiously as possible, without a C-collar and in the left lateral position to definitive care .
Of course, a mature EMS should be able to deliver definitive care in the field. Provision of advanced care such as pre-hospital RSI, inotropic support and blood may afford not just a patent and protected airway, but also the neuroprotective benefits of normoxia, normocarbia and normotension. In contrast to Sundstrom’s proposal of transporting unintubated patients with decreased cosnciousness in left lateral, such patients may be better served by early intubation then be transported supine with measures instituted to minimise spinal movement.
Spinal Immobilisation & The Curse of the ‘Spinal’ Board
Reference to the ‘spinal’ board has been a mainstay of ATLS teaching until very recently. Indeed it is only with the advent of the 9th edition of the ATLS manual that emphasis on the rigid backboard has been swayed from discussion of a ‘spinal protection’ device to an ‘extrication device’ , although many course directors (self included) have purposefully de-emphasised the use of rigid boards for immobilisation for many years prior to approved changes in the ATLS manual. Standard principles of ‘splinting to skin’ and the potential for long transport times (particularly in the rural Australian context) risk the very real danger of pressure necrosis from patients stripped naked and placed on an uncomfortable hard board for transfer to definitive care. Even short periods of immobilisation are associated with consequent pain, a problem worse for patients with altered consciousness or who are intubated and unable to move [28,29]. There is also some evidence that immobilisation on a hard board can cause a reduction in respiratory function [30,31]. For awake patients, there is the potential for motion sickness and vomiting whilst secured supine .
Like the C-collar, rigid boards may still be used in the prehospital setting as an extrication device to facilitate extrication, but have no role to play in spinal immobilisation per se. How then to provide spinal immobilisation in the pre-hospital environment? Supine patients may be best transferred from ground to stretcher using a scoop mattress to minimise movement (bear in mind that patients with clinically significant spinal injuries are at risk of other injuries such as unstable pelvic fractures, thus the dogma of log-rolling on/off a hard board risks worsening both spinal injury, pelvic injury, and skin integrity) .
Newer devices (such as the vacuum mattress or ‘vac mat’) offer both spinal immobilisation from head-to-toe, as well as minimising risk of pressure necrosis . They form a ‘cocoon’ for both conservation of heat and minimising spillage of bodily fluids. Transfer from scoop to a vac mat should preferably occur with a pelvic binder already in situ if there is clinical suspicion of pelvic disruption. If there is concern regarding clinically significant movement of the C-spine despite vac mat, then additional measures such as head blocks may be used (avoiding C-collars due to potential for raised ICP) .
Thus both the traditional dogma of immobilisation on a rigid ‘spinal’ board and universal application of a C-collar to all patients with suspected spinal injuries are being questioned by many [35,36]. Instead careful patient selection and use of appropriate immobilisation is advocated.
Downloadable as a PDF – TRAUMA FLOW CHART
A Pragmatic Approach for Spinal Immobilisation of Pre-Hospital Trauma Patients?
Trauma medicine and pre-hospital care is characterised by the need to make decisions based on a paucity of evidence, making decisions based on the best available evidence and adjusting over time.
One criticism of many of the published reviews of the use of C-collars is that many, whilst explaining the problems of routine C-collar use and documented harms, fail to recommend a practical approach to spinal immobilisation for the future – instead opting for motherhood statements regarding the need for further trials [2,3,4,5,6,7].
Whilst data from large scale randomised controlled trials is preferred, there needs to be careful consideration to the appropriateness of universal C-collar application to trauma patients by EMS providers, mindful of the fact that C-collars are not without risk. Individual EMS may wish to develop their own protocols, under a clinical governance program, to tailor spinal immobilisation strategies according to skillmix of providers, of available devices, of casemix and accounting for transport times . Indeed some EMS have abandoned the use of semi-rigid collars, guided by both literature showing absence of benefit and potential harms, as well as expert opinion.
The Queensland Ambulance Service has recently made a move to replace hard collars with soft, which may be appropriate for both awake and unconscious patients facing prolonged transport times . In contrast, as proposed by Sundstrom recently, Norway EMS providers are moving to abandon routine use of C-collars and transport unconscious, unintubated patients in the left lateral position .
Development of protocols, guided by expert consensus, allows the opportunity for large multi-centre randomised controlled trials to guide future refinements. In the meanwhile a practical approach to spinal immobilisation may be considered as below – using C collars and rigid boards purely to facilitate extrication
Thus best evidence is that C-collars at best serve as a marker that a patient has a potential cervical spine injury. The routine application of a C-collar is inappropriate in isolated penetrating trauma and in patients who are alert, cooperative and can self-maintain neck immobilisation in the absence of a distracting injury. use of NEXUS and Canadian C spine criteria may be helpful in initial assessment by trained providers.
Collars can cause potential harm, so for those patients for whom a cervical spine injury is suspected it may be appropriate to consider manual immobilisation (subject to ergonomics and availability of expertise) or alternative measures including use of a C-collar purely during the extrication phase.
Subsequent management may include the use of soft cervical collars for both awake patients (if desired by patient) and unconscious patients (although the latter may benefit from the use of head blocks to prevent the head from ‘lolling’). In a subset of those unconscious patients for whom airway protection has not been achieved or unavailable, consideration should be made towards transport in the left lateral position, minimising spinal movement.
Rigid boards should only be used as an extrication device, with preference given to minimal handling (avoid log-rolling) using scoop stretcher to lift onto a stretcher to allow complete spinal immobilisation via use of a vacuum mat, supported by pelvic immobilisation if indicated.
- American College of Surgeons Committee on Trauma. Advanced Trauma Life Support – Student Course Manual 9th Edition 2012, American College of Surgeons, Chicago, Illinois
- Abram, S., and Bulstrode, C. Routine spinal immobilization in trauma patients: what are the advantages and disadvantages? Surgeon 2010 8; 218–222
- Deasy, C., and Cameron, P. Routine application of cervical collars—what is the evidence? Injury 2011 42; 841–842
- Oteir A, Smith K, Stoelwinder J, Middleton J & Jennings P Should suspected cervical spinal cord injury be immobilised? Injury 2014 12; 32 doi:10.1016/j.injury.2014.12.032
- Hauswald, M., and Braude, D. Spinal immobilization in trauma patients: is it really necessary? Curr. Opin. Crit. Care 2002 8; 566–570
- Deasy, C., and Cameron, P. Routine application of cervical collars—what is the evidence? Injury 2011 42; 841–842
- Benger, J., and Blackham, J. Why do we put cervical collars on conscious trauma patients? Scand. J. Trauma Resusc. Emerg. Med. 2009 17; 44-48
- Craig, G.R., and Nielsen, M.S. Rigid cervical collars and intracranial pressure. Intensive Care Med. 1991 17; 504–505
- Davies, G., Deakin, C., and Wilson, A. The effect of a rigid collar on intracranial pressure. Injury 1996 27; 647–649
- Hunt, K., Hallworth, S., and Smith, M. The effects of rigid collar placement on intracranial and cerebral perfusion pressures. Anaesthesia 2001 56; 511–513
- Criswell, J.C., Parr, M.J., and Nolan, J.P. Emergency airway management in patients with cervical spine injuries. Anaesthesia 1994 49; 900–903
- Doran, J.V., Tortella, B.J., Drivet, W.J., and Lavery, R.F. Factors influencing successful intubation in the prehospital setting. Prehosp. Disaster Med. 1995 10; 259–264
- Ghafoor, A.U., Martin, T.W., Gopalakrishnan, S., and Viswamitra, S. Caring for the patients with cervical spine injuries: what have we learned? J. Clin. Anesth. 2005 17; 640–649
- Goutcher, C.M., and Lochhead, V. Reduction in mouth opening with semi-rigid cervical collars. Br. J. Anaesth. 2005 95; 344–348
- Robitaille, A. Airway management in the patient with potential cervical spine instability: continuing professional development. Can. J. Anaesth. 2011 58; 1125–1139
- Hauswald, M., Ong, G., Tandberg, D., and Omar, Z. Out-of-hospital spinal immobilization: its effect on neurologic injury. Acad. Emerg. Med. 1998 5; 214–219
- Thumbikat, P., Hariharan, R.P., Ravichandran, G., McClelland, M.R., and Mathew, K.M. Spinal cord injury in patients with ankylosing spondylitis: a 10-year review. Spine 2007 32; 2989–2995
- Kosiak M: Etiology of decubitus ulcers. Arch Phys Med Rehabil 1961, 42; 19-29
- Patterson RP, Cranmer HH, Fisher SV, Engel RR: The impaired response of spinal cord injured individuals to repeated surface pressure loads. Arch Phys Med Rehabil 1993, 74; 947-953
- Sporer, K.A. Why we need to rethink C-spine immobilization: we need to reevaluate current practices and develop a saner cervical policy. EMS World 2012 41; 74–76
- International Liaison Committee On Resuscitation. Draft statement: Among adults and children with suspected traumatic cervical spinal injury (P), does spinal motion restriction (I), compared with no spinal motion restriction (C), change neurological injury, complications, overall mortality, pain, patient comfort, movement of the spine, hospital length of stay (O)? February 2015. Available via URL https://volunteer.heart.org/apps/pico/Pages/PublicComment.aspx?q=772
- Kwan, I., Bunn, F., and Roberts, I.; WHO Pre-Hospital Trauma Care Steering Committee. Spinal immobilisation for trauma patients.Cochrane Database Syst. Rev. Issue 2001 2. Art. No.: CD002803
- Stiell IG, Clement CM, McKnight RD The Canadian C-spine rule versus the NEXUS low-risk criteria in patients with trauma. N Engl J Med 2003; 349:2510–2518
- Weingart S. A pain in the neck. EMCrit Podcast No. 63 2011 Available via URL http://emcrit.org/podcasts/cervical-spine-injuries-i/
- Lockey, D.J., Coats, T., and Parr, M.J. Aspiration in severe trauma: a prospective study. Anaesthesia 1999 54; 1097–1098
- Delrossi G, DuBose D, Scott N, Conrad B, Hyldmo P, Rechtine G & Horodyski M. Motion Produced in the Unstable Cervical Spine by the HAINES and Lateral Recovery Positions. Prehospital Emerg Care 2014 18(4); 539-543
- Sundstrøm T, Asbjørnsen H, Habiba S, Sunde GA, Wester K. Prehospital Use of Cervical Collars in Trauma Patients: A Critical Review. J Neurotrauma. 2015 Epub ahead of print.
- Main PW, Lovell ME: A review of 7 support surfaces with emphasis on their protection of the spinally injured. J Accid Emerg Med 1996 13; 34-37
- Barney RN, Cordell WH, Miller E: Pain associated with immobilization on rigid spine boards. Ann Emerg Med 1989 18; 918
- Bauer D, Kowalski R: Effect of spinal immobilization devices on pulmonary function in the healthy, nonsmoking man. Ann Emerg Med 1983 17(9); 915-918
- Totten, V.Y., and Sugarman, D.B. Respiratory effects of spinal immobilization. Prehosp. Emerg. Care 1999 3; 347–352
- Krell JM, McCoy MS, Sparto PJ et al. Comparison of the Ferno Scoop Stretcher with the long backboard for spinal immobilization. Prehosp Emerg Care 2006 10; 46-51
- Luscombe M, & Williams J Comparison of a long spinal board and vacuum mattress for spinal immobilisation EMJ 2003 20(5); 476-478
- Holla, M. Value of a rigid collar in addition to head blocks: a proof of principle study. Emerg. Med. J. 2012 29; 104–107
- Dolven T Cervical collar RIP. 2014. Available via URL http://www.scancrit.com/2014/04/02/cervical-collar-r-i-p/
- Le Cong, M. So you gave up cricoid pressure but not the cervical collar. 2014. Available via URL http://prehospitalmed.com/2014/04/03/so-you-gave-up-cricoid-pressure-but-not-the-cervical-collar/